Morphological changes in the radula of abalone Haliotis diversicolor aquatilis from post-larva to adult.

ABSTRACT The radula morphology of Haliotis diversicolor aquatilis was examined by SEM from postlarval to adult stages. In postlarvae of -0.5 mm SL, a rachidian tooth (R), 2 pairs of lateral teeth (L1, L2), and 2-3 pairs of marginal teeth (M) per row were present. Only the number of marginal teeth increased to 12-13 pairs per row by 1.5 mm SL. An additional 2 pairs of lateral teeth (L3, L4) were added progressively from 1.5-2 mm SL. The L5 teeth were added at about 6 mm SL. Postlarvae with <2 mm SL had strongly curved rachidian and lateral teeth with clearance angles of -15[degrees] to 20[degrees]. Each tooth had many serrations on the working edges. These structures of the teeth are probably suitable for collecting biofilm components such as extracellular mucus of diatoms. Postlarvae with >2.5 mm SL had higher clearance angles of 20[degrees] to 35[degrees]. Serrations on the working edges of rachidian and lateral teeth became progressively shallower. These changes suggest that the teeth are more suitable for grazing substrata to efficiently remove strongly attached diatoms. As abalone grew larger than 2 mm SL, the L3 and L4 teeth became longer than R, L1, and L2. The radula development of H. diversicolor aquatilis shown in this study is similar to that of H. discus hannai. However, it is suggested that the feeding transitions in the postlarval stage occur later in H. diversicolor aquatilis than in H. discus hannai. The morphology of juvenile and adult radula teeth of H. diversicolor aquatilis is clearly distinguishable from that of H. discus hannai. The sizes of L4 and L5 teeth of juvenile and adult H. diversicolor aquatilis were much smaller than L3 teeth, whereas sizes of L3-L5 teeth were not clearly different in H. discus hannai. This difference in radula morphology in juvenile and adult stages suggests that the main food items for juvenile and adult H. diversicolor aquatilis may be different from those of H. discus hannai.

KEY WORDS: abalone, radula, development, morphological changes, feeding, Haliotis diversicolor

INTRODUCTION

Haliotis diversicolor aquatilis Reeve 1846 is a commercially important abalone in Japan. However, the catch of this species has been declining probably because of over-fishing. Because reseeding of hatchery-reared juveniles into wild populations has been conducted in many places to enhance the stocks, ecologic studies of this species have been very limited, especially in the early life stages. Since 2001, we have started studying the reproduction and early life ecology of H. diversicolor aquatilis at Nagai in Sagami Bay, Japan. Natural recruitment of this species at Nagai seemed relatively high in comparison with those of other abalone species (H. discus discus Reeve 1846, H. madaka Habe 1979 and H. gigantea Gmelin 1791) in the same area, but growth and survival rates of postlarval H. diversicolor aquatilis was variable among sampling areas (Onitsuka et al. unpublished).

Quantity and quality of food are important factors affecting the growth and survival of postlarval abalone (Kawamura et al. 1998), and these factors have been suggested to be critical in controlling recruitment of H. discus hannai Ino 1953 (Kawamura et al. 2002). Feeding habits in the early life stages of abalone have been studied in detail for H. discus hannai in recent years (reviewed by Kawamura et al. 2002, Takami & Kawamura 2003), and three major transitions in feeding have been identified (Kawamura et al. 1998, Takami & Kawamura 2003). The transitions in feeding are suggested to be closely related to structural and functional changes in the radula during postlarval and juvenile stages (Kawamura et al. 2001). However, there is little information concerning the feeding habits of H. diversicolor aquatilis, and the diets in the early life stages are generally unknown, as well as for other many abalone species. In addition, the radula morphology of H. diversicolor aquatilis has only been examined for adults (Wu 1991). In the present study, changes in radula morphology of H. diversicolor aquatilis are described from postlarva to adult. By comparing the radula development with that of H. discus hannai (Kawamura et al. 2001) we try to estimate the feeding habits of H. diversicolor aquatilis in postlarval and juvenile stages.

MATERIALS AND METHODS

Abalone Rearing

Larval H. diversicolor aquatilis were hatched in October 2000, September 2001, and August 2002 at the Nagasaki Prefectural Institute of Fisheries (Nagasaki, Japan). One-day-old larvae were transferred to culture tanks with plastic plates covered with naturally appeared benthic diatoms as settlement plates. Postlarvae that settled on the plates were reared continuously in sand-filtered running seawater at 20[degrees]C.

In this study, we defined the boundary between postlarval and juvenile H. diversicolor aquatilis as the size at which the first respiratory pore is sealed (~2.5 mm SL, Onitsuka et al. unpublished), following the definition for H. discus hannai by Takami (2002). We call individuals larger than 35 mm SL adults according to Oba (1964).

Radala Observation

Samples of postlarvae and juveniles were preserved in 5% formalin in seawater at intervals of 3 days from 10 to 25 days postfertilization, and at 55 days and 163 days postfertilization in 2002. One, 3, 4, and 16-mo-old juveniles in 2001, 28-mo-old juveniles in 2000 were also preserved in 5% formalin in seawater. Radulae of postlarvae and small juveniles were removed with a pipette under an inverted microscope following dissolution of tissues by soaking in sodium hypochlorite (0.6% CI concentration; Wako Pure Chemical Industries Ltd, Osaka, Japan) for several minutes. Radulae were then serially pipetted through several distilled water baths to remove residual sodium hypochlorite. This procedure has been shown to preserve the structure of radula teeth (Kawamura et al. 2001). Abalone size represented by the longest shell length (SL) was measured individually before dissolution. Radulae of large juveniles and adults were dissected from preserved animals under a dissecting microscope after measuring shell length. The radulae were soaked in sodium hypochlorite to dissolve surrounding tissues attached to the radulae and then washed with distilled water to remove residual sodium hypochlorite. Because we could not get adult abalone from the hatchery, wild adult H. diversicolor aquatilis of 52.5-76.9 mm SL were caught at Nagai in 2002 as adult samples.

Radula length, width, number of transverse rows of teeth (Fig. 1A), and number of lateral and marginal teeth per row (except for individuals smaller than 6.5 mm SL) were determined using a monitor and digital camera system with an image analyzer, connected to an inverted microscope and a dissecting microscope. Radulae were then transferred to scanning electron microscope (SEM) stubs, laid flat with the teeth upward, and allowed to air dry before sputter coating with platinum for SEM observations.

[FIGURE 1 OMITTED]

The length and width of rachidian teeth, lengths of the lateral teeth, gap between the rachidian teeth of adjacent transverse rows (Fig. 1B), and number of lateral and marginal teeth per row were measured using SEM photographs. Relative length of L4 and L5 teeth to L3 teeth was also calculated. The clearance angle of rachidian and lateral teeth (L1, L2; Fig. 1C, D) of postlarval and juvenile (<10 mm) radula, which was suggested to provide information on the function of radula teeth (Padilla 1985), was measured using the method by Roberts et al. (1999b). These measurements were done for five teeth (R, L1, L2) on a row located in the middle part of the whole radula. The clearance angles of older juvenile and adult radula were not measured because accurate measurements were obstructed by the overhanging marginal teeth. Measurements of the angles of rachidian teeth and inner lateral teeth of juveniles and adults could not be performed because they were located behind the lengthened outer laterals.

RESULTS

During the postlarval and juvenile period the radula underwent various morphologic developments before reaching the adult form. Most developments correlated more strongly with shell length than age (Table 1), so the following description relates primarily to shell length.

Overall radula length increased linearly with shell length (Fig. 2). However, the factors determining length changed as the abalone developed. During the early post larval period (<~0.8 mm SL), the number of transverse rows of teeth was the main factor for increasing radula length. The radula of postlarvae with ~0.5 mm SL contained 19 to 21 rows of teeth, and the number of rows increased to 22 to 26 at ~0.8 mm SL (Fig. 3B). Late postlarvae (0.8-2 mm) had 22 to 26 rows of teeth, therefore the increase in radula length reflected an increased gap between adjacent rows of teeth (Fig. 2) and an increased length of rachidian and lateral teeth (indicated by rachidian and L3 in Fig. 2). The number of rows of teeth started increasing again above a size of 2 mm (Figs. 3A, B). Above 1.5 mm SL the gap between rows of teeth increased rapidly (Fig. 2) as differentiation of the L3-L5 teeth developed. Therefore, the increase in radula length of abalone >2 mm SL was caused by increases in row number (Figs. 3A, B), the gap between rows of teeth and teeth length (Fig. 2).

[FIGURES 2-3 OMITTED]

The width of the radula also increased linearly as abalone grew (Fig. 4). The increase in radula width for postlarvae was mainly caused by the increase in the number of marginal (Fig. 5) and lateral teeth per row (Fig. 6). Postlarvae with ~0.5 mm SL contained the rachidian tooth (R), 2 pairs of lateral teeth (L1 and L2), and 2 to 3 pairs of marginal teeth per row (Fig. 7A), although lateral and marginal teeth were not clearly differentiated. The number of marginal teeth per row increased progressively; postlarvae with shell lengths of 1.5, 6, and [greater than or equal to] 30 mm had 12-13, 30-40, and 70-80 pairs of teeth respectively (Fig. 5). Postlarvae smaller than 1.5 mm SL contained only 2 pairs of lateral teeth (L1 and L2) (Fig. 7B). Two additional pairs of lateral teeth (L3 and L4) were added progressively in animals with ~1.5 mm to ~2 mm SL (Fig. 6). A pair of L5 was added in animals with about 6 mm SL, and at this size the adult complement of lateral teeth was completed (Fig. 6). The increase in width during juvenile stage was caused mainly by the increase in width of the individual teeth (illustrated by the width of the rachidian teeth in Fig. 4) and by the increase in the number of marginal teeth per row (Fig. 5). The increase in radula width of adults was caused by the increase in width of the individual teeth (Fig. 4).

[FIGURES 4-7 OMITTED]

Changes in the Morphology of Individual Teeth

The serrations on the working edges of the rachidian and lateral teeth became less pronounced as abalone grew. R, L1, and L2 teeth initially had long, pointed serrations (Figs. 7A, B). In animals with more than 1 mm SL, these serrations became progressively shallower, first on R, and later on L1 and L2 as L3-L5 developed. By approximately 2 mm SL, nearly all serrations had disappeared from R, L1, and L2 teeth, but L3 and L4 retained serrations on their edges (Figs. 7D, E). All serrations disappeared from L3-L5 teeth in adult radula (Fig. 7F).

The L3-L5 teeth were already slightly longer than R, L1, and L2 when they were added, but difference became greater as abalone grew (Figs. 7D-F, Fig. 8A). This differentiation of the lateral teeth accompanied an increase in the gap between adjacent rows of teeth (Fig. 2). The L3-L5 teeth became pointed, similar to canine teeth, whereas the tip of R and L1 teeth became flat, similar to spades (Fig. 7F), during the juvenile and adult stages. The L2 teeth of adults were longer and much more pointed at the tip than R and L1, but much shorter than L3. The sizes of outer lateral teeth (L3-L5) were different in juvenile and adult radula. The relative lengths of L4 and L5 teeth to L3 teeth were 0.7 and 0.5 (Fig. 8B), although their shapes were similar (Figs. 7E, F). The size of L5 teeth was almost the same as marginal teeth, but much more pointed at the tip than marginals (Fig. 7F).

[FIGURE 8 OMITTED]

Marginal teeth were narrow and comb-like, with many fine serrations present from postlarva to adult (Fig. 7). The size of individual marginal teeth relative to the rachidian and lateral teeth became smaller, and the tip of marginal teeth became more round as abalone grew. The marginal teeth of adult abalone retained serrations on their edges, but the serrations became less pronounced near the tip (Fig. 7F).

Post-larval abalone smaller than about 2 mm SL had strongly curved rachidian and lateral teeth (Fig. 9A) with mean clearance angles of about -15[degrees] to 20[degrees] (Fig. 10), although the angle was variable within and between radulae. The clearance angle increased in the late postlarval stage between 2-2.5 mm SL, and the rachidian and lateral teeth of juveniles (>2.5 mm SL) had higher clearance angles of 20[degrees] to 35[degrees] (Figs. 9B, 10).

[FIGURES 9-10 OMITTED]

DISCUSSION

Morphology of Adult Radula

Kawamura et al. (2001) showed that the development of the radula of H. discus hannai was remarkably similar to that of H. iris Gmelin 1791 (Roberts et al. 1999b). The morphology of adult teeth was not clearly distinguishable between the two species. These two species are not genetically close (Lee & Vacquier 1995), but their feeding habits and growth patterns are similar (Kawamura et al. 1998). The radula teeth of adult H. diversicolor aquatilis, shown in this study, were easily distinguished from the radula of H. iris (Roberts et al. 1999b) and H. discus hannai (Kawamura et al. 2001). The size of outer lateral teeth (L3-L5) was apparently different in H. diversicolor aquatilis, L3 was the largest and L5 was the smallest (Figs. 7D-F, 8), while sizes of L3-L5 teeth were not clearly different in H. iris and H. discus hannai. Similar differences in the size of L3-L5 teeth were seen in tropical small abalone species 14. asinina Linnaeus 1758, H. ovina Gmelin 1791 (Chitramvong et al. 1998), and H. varia Linnaeus 1758 (Wu 1991); but greater differences were observed in H. diversicolor aquatilis. The different morphologies of adult radula teeth may indicate different feeding habits among abalone species.

Development Process of Radula

The development process of the radula in H. diversicolor aquatilis was similar, but generally slower than in H. iris (Roberts et al. 1999b) and H. discus hannai (Kawamura et al. 2001). The initial increase in the number of teeth rows after metamorphosis was observed until the animal reached ~0.8 mm SL in H. diversicolor aquatilis (Fig. 3B), whereas in H. iris (Roberts et al. 1999b) and H. discus hannai (Kawamura et al. 2001) it was until ~0.5 mm SL. However, the size at which the row number started increasing again was ~2 mm SL in H. diversicolor aquatilis, which is smaller than the size of the animal with 4-5 mm SL reported for H. discus hannai (Kawamura et al. 2001). The mean postlarval size (the longest dimension of the larval shell) at metamorphosis of H. diversicolor aquatilis (257 [micro]m) is smaller than that of H. discus hannai (277 [micro]m) (Hayashi 1983). The remarkable increase in clearance angle of rachidian and lateral teeth was observed at 2-2.5 mm SL in H. diversicolor aquatilis (Fig. 10), whereas it is noticed at ~1 mm SL in H. iris (Roberts et al. 1999b) and H. discus hannai (Kawamura et al. 2001). The L5 teeth were added completing the adult complement of five pairs of laterals at ~6 mm SL in H. diversicolor aquatilis (Fig. 6), whereas at ~1.7 mm SL in H. iris (Roberts et al. 1999b) and at ~2 mm SL in H. discus hannai (Kawamura et al. 2001).

Estimated Transitions in Feeding

Three major transitions in feeding have been identified in H. discus hannai (Kawamura et al. 1998, Takami & Kawamura 2003), which appears to be closely related to radula development (Kawamura et al. 2001). The first transition is from lecithotrophy to particle feeding, around the time of metamorphosis. Postlarval H. iris (Roberts et al. 1999a) and H. discus hannai (Seki & Kan-no 1981, Kawamura & Takami 1995) begin feeding, within a day of the velum being shed, on the surface of crustose coralline algae (CCA) where larval abalone preferentially settle in the natural environment (e.g., Saito 1981, Morse & Morse 1984). However, postlarvae can grow up to ~0.4 mm SL without food during several days following metamorphosis (Roberts et al. 2001, Takami et al. 2000) by using their residual yolk supply as an energy source (Roberts et al. 2001, Takami et al. 2002). The early increase in the number of teeth rows was suggested to be related to the first transition in feeding, indicating active formation of new rows of teeth prior to shedding anterior worn teeth during feeding (Kawamura et al. 2001). The greater size at which H. diversicolor aquatilis concluded the initial addition of teeth rows (0.8 mm SL) may indicate longer contribution of residual yolk reserves as an energy source after metamorphosis for this species. After the initial increase, the number of rows remained relatively constant until ~2 mm SL and then started increasing again (Fig. 3B). In H. discus hannai the number of rows was relatively constant between ~0.5 mm and 5 mm SL (Kawamura et al. 2001), which is longer than in H. diversicolor aquatilis in the present study. The factors controlling the increasing number of rows are not known, but this may be an interesting difference between the two abalone species.

[FIGURE 3 OMITTED]

Clearance angles of zero or less were suggested to indicate teeth would slide across the surface rather than cutting it (Padilla 1985). The curved radula teeth of postlarval H. diversicolor aquatilis <2 mm SL are probably suitable for collecting biofilm materials such as extracellular mucus secretions of diatoms, as suggested for postlarvae <1 mm of H. iris (Roberts et al. 1999b) and H. discus hannai (Kawamura et al. 2001). The increase in clearance angle in H. discus hannai of approximately 1 mm SL appeared to be related to the second transition in feeding, main diets being from mucous materials to tightly attached diatoms such as Cocconeis spp. colonizing on CCA (Kawamura et al. 2001). Postlarval H. discus hannai >1 mm SL with positive clearance angles of radula teeth seem able to detach Cocconeis cells from CCA surfaces, whereas smaller postlarvae with curved radula teeth cannot efficiently detach diatom cells (Kawamura et al. 2001). H. diversicolor aquatilis appears to efficiently detach strongly attached diatoms from CCA surfaces at 2-2.5 mm SL when the radula teeth have higher clearance angles. Biofilm components such as extracellular secretions from diatoms and/or from CCA can be important food sources for postlarval H. diversicolor aquatilis <2 mm SL, which is larger than in H. discus hannai (<1 mm SL). However, the size of food items that can be collected by the radula seems to increase with a reduction in tooth serrations (1 mm SL, Fig. 7) and an increase in the gap between the adjacent rows of teeth of postlarvae (>1.5 mm SL, Fig. 2). It should be noted here that the clearance angle at the point of contact with food items can change depending on various factors, such as the position of the radula when it protrudes for feeding, flexing of the radula during feeding, and the nature of the substrate. Our clearance angle data are based on the radulae removed from abalone and prepared for SEM. Therefore, the above discussion on the implications of feeding due to increased clearance angles is conjectural, as discussed previously in Roberts et al. (1999b) and Kawamura et al. (2001).

[FIGURE 7 OMITTED]

The L3-L5 teeth of juveniles and adults were sharply cusped without serrations and stood much higher above the radula membrane than rachidian and L1 and L2 teeth (Figs. 7D-F), suggesting that they are suitable for cutting firmly attached algae and/or for excavating substrata, as discussed for H. iris (Roberts et al. 1999b) and H. discus hannai (Kawamura et al. 2001). The differentiation of the L3-L5 teeth was considered to be related to the final transition in feeding of H. discus hannai, from microalgal to macroalgal feeding (Kawamura et al. 2001), though the marked increase in the activity of the macroalgal polysaccharide degrading enzymes observed at about 2 mm SL also appeared to contribute to the efficient utilization of macroalgae (Takami et al. 1998). In fact, postlarval H. discus hannai >1.8 mm SL used juvenile macroalgae L. japonica Areschoug 1851 efficiently (Takami 2002) after the differentiation of lateral teeth, that started at ~1.5 mm SL (Kawamura et al. 2001). Although the adult complement of five pairs of laterals was completed at ~6 mm SL in H. diversicolor aquatilis (Fig. 6), the differentiation of the L3 teeth started at ~1.5 mm SL (Fig. 2) from the time they appeared. This suggests that H. diversicolor aquatilis also starts utilizing juvenile macroalgae in the late postlarval stage. Hence, H. diversicolor aquatilis may not need to use tightly attached diatoms before they start feeding on juvenile macroalgae, because they can feed and grow on biofilm mucus materials until ~2 mm SL, as discussed earlier.

[FIGURE 6 OMITTED]

Juveniles of Haliotis discus hannai and its subspecies H. discus discus initially feed on juvenile macroalgae (Maesako et al. 1984, Takami 2002) and/or soft algae such as Ulva spp. (Seki 1997). They began feeding on adult macroalgal fronds such as Laminaria and Eisenia once they became larger than ~13 mm SL (Seki 1997). The morphology of juvenile and adult radula teeth of H. diversicolor aquatilis is different from that of H. discus hannai. The sizes of L4 and L5 teeth of juvenile and adult H. diversicolor aquatilis were much smaller than L3 teeth (Figs. 7D-F, Fig. 8). Although we do not know how the abalone use different types of the radula teeth during feeding, the difference in radula morphology suggests that the main food items for juvenile and adult H. diversicolor aquatilis are different from H. discus hannai.

[FIGURE 8 OMITTED]

By comparing the radula morphology and its development with those of H. discus hannai, we can estimate the transitions in feeding of H. diversicolor aquatilis as earlier mentioned. Three possible differences in feeding and nutritional sources between the two species, which may be important from the ecologic point of view, can be considered: (1) The longer contribution of the residual yolk reserve as an energy source after metamorphosis is suggested for H. diversicolor aquatilis. This may mean longer starvation tolerance of newly metamorphosed postlarvae with limited and/or unsuitable food sources, or longer swimming larval life if they fail to contact an appropriate environmental stimulus for metamorphosis than those of H. discus hannai (Takami et al. 2000, Takami et al. 2002). (2) Because the differentiation of outer lateral teeth started almost at the same time of increasing clearance angles, main food items for H. diversicolor aquatilis can possibly transfer directly from biofilm mucus materials to juvenile macroalgae, and tightly attached diatoms may not be an essential food source in between them. Competition for tightly attached diatoms on CCA habitats between other herbivore gastropods was suggested to be a critical factor affecting the survival and growth rate of H. discus hannai (Takami et al. 2001, Kawamura et al. 2003). If the diatoms would not be necessary, the feeding and growth of H. diversicolor aquatilis could be more flexible than H. discus hannai. (3) The difference in the main food items for juveniles and adults between the two species, suggested in this study, may reflect differences in their actual habitats. The adults H. diversicolor aquatilis inhabit shallow boulder areas where large macroalgal forests are limited, whereas adult H. discus hannai are in deeper rocky areas with macroalgal forests. Further field observations and laboratory experiments are needed to clarify the feeding and nutritional sources of H. diversicolor aquatilis throughout their life stages.

TABLE 1.

Correlation of radula variables with post-larval age and post-larval
shell length. Post-larvae were chosen to include a range of sizes for
each age. Data are Pearson's correlation.

              Variable                 Shell Length    Age    n

Overall length of radula                  0.998       0.953   76
Overall width of radula                   0.992       0.964   76
Number of rows of teeth on radula         0.979       0.943   76
Width of rachidian tooth                  0.971       0.901   31
Length of rachidian tooth                 0.937       0.899   31
Gap between adjacent rachidian teeth      0.977       0.944   31
Number of lateral teeth per row           0.934       0.886   31

ACKNOWLEDGMENTS

The authors thank Christopher B. Clarke for critical reading of a draft, and M. Hara for help with SEM observations.

LITERATURE CITED

Chitramvong, Y. P., E. S. Upatham, M. Kruatrachue, P. Sobhon & V. Limthong. 1998. Scanning electron microscope study of radulae in Haliotis asinina Linnaeus, 1758 and Haliotis ovina Gmelin, 1791 (Gastropoda: Haliotidae). J. Shellfish Res. 17:755-759.

Hayashi, I. 1983. Larval shell morphology of some Japanese Haliotids for the identification of their veliger larvae and early juveniles (in Japanese with English abstract). Venus 42:49-58.

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. Fisheries Sci. 61:357-358.

Kawamura, T., H. Takami & Y. Yamashita. 2003. Effect of grazing by a herbivorous gastropod Homalopoma amussitatum, a competitor for food with post-larval abalone, on a community of benthic diatoms. J. Shellfish Res. 23:000-000.

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

Kawamura, T., H. Takami & T. Saido. 2002. Early life ecology of abalone Haliotis discus hannai in relation to recruitment fluctuations. Fisheries Sci. 68(suppl. 1):230-234.

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 Sci. 67:596-605.

Lee, Y.-H. & V. D. Vacquier. 1995. Evolution and systematics in Haliotidae (Mollusca: Gastropoda): inferences from DNA sequences of sperm lysin. Mar. Biol. 124:267-278.

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 Proof. Inst. Fish. 10:53-56.

Morse, A. N. C. & D. E. Morse. 1984. Recruitment and metamorphosis of Haliotis larvae induced by molecules uniquely available at the surface of crustose red algae. J. Exp. Mar. Biol. Ecol. 75:191-215.

Oba, T. 1964. Study on the propagation of an abalone, Haliotis diversicolor supertexta Lischke-1 (in Japanese with English abstract). Bull. Jap. Soc. Fish. 30:742-748.

Padilla, D. K. 1985. The structural resistance of algae to herbivores: a biomechanical approach. Mar. Biol. 90:103-109.

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

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

Roberts, R. D., C. Lapworth & R. Barker. 2001. Effect of starvation on the growth and survival of post-larval abalone (Haliotis iris). Aquaculture 200:323-338.

Saito, K. 1981. The appearance and growth of 0-year-old Ezo abalone. Bull. Jap. Soc. Fish. 47:1393-1401.

Seki, T. 1997. Biological studies on the seed production of the northern Japanese abalone, Haliotis discus hannai Ino (in Japanese with English abstract). Bull. Tohoku Reg. Fish. Res. Lab. 59:1-71.

Seki, T. & H. Kan-no. 1981. Observations on the settlement and metamorphosis of the veliger of the Japanese abalone, Haliotis discus hannai Ino, Haliotidae, Gastropoda (in Japanese with English abstract). Bull. Tohoku Reg. Fish. Res. Lab. 42:31-39.

Takami, H. 2002. Studies on the feeding, growth and survival in early life stages of an abalone Haliotis discus hannai (in Japanese). Doctoral thesis, Tokyo: The University of Tokyo. 220 pp.

Takami, H. & T. Kawamura. 2003. Dietary changes in the abalone, Haliotis discus hannai, and relationship with the development of digestive organ. Japan Agricultural Research Quarterly 37:91-100.

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., T. Kawamura & Y. Yamashita. 2000. Starvation tolerance of newly metamorphosed abalone Haliotis discus hannai. Fisheries Sci. 66:1180-1182.

Takami, H., T. Kawamura & Y. Yamashita. 2002. Effects of delayed metamorphosis on larval competence, and postlarval survival and growth of abalone Haliotis discus hannai. Aquaculture 213:311-322.

Takami, H., T. Kawamura, H. Ito, M. Seito, S. Yanagiya & Y. Yamashita. 2001. Competition for food between post-larval abalone and two species of Herbivorous Gastropods (in Japanese with English abstract). Bull. Tohoku Natl. Fish. Res. Inst. 64:43-51.

Voltzow, J. 1994. Gastropoda: Prosobranchia. In: F. W. Harrison & A. J. Kohn, editors. Microscopic anatomy of invertebrates, Vol. 5, Mollusca I. New Yolk: Wiley-Liss Inc. pp. 111-252.

Wu, W. L. 1991. SEM comparison of radulae in six Taiwanese archaeogastropods. Bull. Inst. Zool. Academia Sinica. 30:299-310.

TOSHIHIRO ONITSUKA, (1) * TOMOHIKO KAWAMURA, (1) SATOSHI OHASHI, (2) TOYOMITSU HORII, (3) AND YOSHIRO WATANABE (1)

(1) Ocean Research Institute, The University of Tokyo, Nakano, Tokyo 164-8639, Japan; (2) Nagasaki Prefectural Institute of Fisheries, Nagasaki, Nagasaki 851-2213, Japan; (3) National Research Institute of Fisheries Science, Yokosuka, Kanagawa 238-0316, Japan

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