Growth responses in emergent greenlip abalone to density reductions and translocations.

ABSTRACT Growth of "stunted" greenlip abalone in areas with low maximum sizes was enhanced over 6 mo by reducing their natural density and by translocation to habitats supporting faster growing abalone. Density reductions significantly increased growth relative to controls, apparently without altering the asymptotic length. Stunted abalone showed a consistent and similar pattern of enhanced growth when translocated to two sites where abalone characteristically grow taster and to larger sizes. When compared with slow growth control abalone, the response of translocated abalone varied with initial length in the same manner as in the experiment where density was reduced. When compared with fast growth controls, translocated abalone had similar trends in growth increment versus size, yet all size categories grew consistently less. Statistical comparisons cannot be made between density-reduced and translocated abalone because abalone in the density reduction experiments were tagged in situ, whereas translocated abalone were tagged aboard a research vessel. The fact that reduced density and better quality habitat positively influence growth patterns of greenlip abalone, producing the same short-term response, suggests that food availability may limit the growth of stunted populations. The asymptotic length, which appeared unaffected in both experiments, may be determined by long-term conditions, or perhaps by conditions during the onset of maturity.

KEY WORDS: abalone, Haliotis laevigata, density-dependence, growth, density-reduction, translocation

INTRODUCTION

An understanding of density dependent processes is essential to fisheries management, particularly compensatory effects that can offset natural or fishing-induced reductions in biomass (Rose et al. 2001). Several studies have reported interspecific competition that resulted in density dependent changes in growth rate (Kamermans et al. 1992, Brazeiro & Defeo 1999, Barki et al. 2001, Talman & Keough 2001, Vromant et al. 2002), but experimental manipulations of density to determine effects of intraspecific competition in the wild are often logistically difficult, particularly for subtidal species.

Stoner (1989) showed that growth of the queen conch, Strombus gigas, was significantly different at densities two and four times those observed in the wild and he suggested that this result was food related. Marshall and Keough (1994) also provided evidence of density-limited growth for the limpet, Cellana tramoserica. Abalone, like conch and limpets, can often be found in high-density aggregations (Shepherd 1986, McShane 1995, Officer et al. 2001), and because they feed on drift algae it is reasonable to postulate that abalone density, as a result of intraspecific competition for food, may affect growth.

Most abalone research investigating density dependent growth has been restricted to aquaculture experiments (Hunt et al. 1995, Mgaya & Mercer 1995, Capinpin Jr. et al. 1999, Huchette et al. 2003), and in each case growth was negatively correlated with density. In contrast to these results, McShane and Naylor (1995a) reported density independent growth of Haliotis iris from field experiments in New Zealand. In their experiment McShane and Naylor artificially increased densities within enclosures, using natural densities for the controls. They acknowledged that a more realistic experiment to test density effects would be achieved with a reduction, rather than increase, in density. "Stunted" populations of abalone (McShane et al. 1994, Wells & Mulvay 1995), where few individuals will ever grow to the legal minimum length and therefore most will remain unfished, provide a unique opportunity to conduct such manipulative experiments in a population at natural density.

Translocation of abalone to habitats with greater food abundance or quality has been shown to increase growth rate. Emmett and Jamieson (1989) transplanted stunted H. kamtschatkana to better quality habitats and obtained growth rates twice those of control populations. McShane and Naylor (1995b) also observed significant increases in growth rate when H. iris were translocated from bay to headland habitats. A review by Day and Fleming (1992) concluded that variation in abalone growth was primarily due to the quality and quantity of algal food available. Assuming that density affects food availability per individual, a reduction in density could be expected to invoke a similar growth response as translocation of abalone to better quality habitats.

Our experiment, conducted over a 6-mo period, studied the effect of density reductions at three sites with stunted abalone at Tiparra reef, South Australia. Abalone removed from two of these sites were translocated to locations with fast growing abalone, and controls of both stunted abalone and fast growing abalone were tagged to determine the effect of changing habitat on growth.

METHODS

During March and April 2002, 2,940 greenlip abalone were tagged at five sites on Tiparra Reef, South Australia (Fig. 1), using rivet tags inserted into a respiratory pore of the shell (Prince 1991). Tagged abalone (1,873) were recaptured and measured for shell growth during September 2002.

[FIGURE 1 OMITTED]

The experiment to determine the effect of density on shell growth was established at three sites where abalone grew to small maximum lengths. We assumed that growth at these sites would be slow, and the results confirm this. These sites were within a kilometer of each other, separated by areas of unsuitable habitat. Within each site two adjacent areas were marked out and measured, and abalone were tagged in situ in each area. We did not anticipate extensive movements during these experiments because Shepherd (1973) observed that movements were minimal over long periods of time at a site near to the Tiparra Lighthouse. We therefore established the control areas only 10 m from treatment areas to minimize differences between them. In control areas all emergent abalone were tagged, and in treatment areas every third emergent abalone encountered was tagged, during systematic searches of the areas. After tagging was completed all untagged emergent abalone in treatment areas were removed. The size of density-reduced areas was approximately three times larger than control areas to ensure similar numbers of tagged individuals after thinning.

Recapture surveys also involved systematic searches of each area, but when aggregations were encountered, the numbers of tagged and untagged abalone were counted. At each site the original density was estimated using the number tagged in March, and the final density in September was calculated from both the number recaptured and the proportion of tagged to untagged abalone in aggregations during the recapture surveys. This latter density estimate assumed that the proportion of tagged individuals within the aggregations measured reflects the proportion of tagged to untagged individuals in the remainder of the area. An aggregation was defined as a group of abalone with no more than 150 cm between two individual abalone, a key distance for fertilization success of H. laevigata (Babcock & Keesing 1999). A high proportion of abalone were in aggregations at all sites. Only abalone found in their area of origin were included in the abundance estimation.

Each site varied in degree of aggregation as well as habitat. The "Lighthouse" site was at 3-m depth, with smaller aggregations of abalone that were relatively evenly distributed on continuous limestone habitat. The "Aggregation" site consisted of small to large aggregations of tightly clustered abalone at 6-m depth. Aggregations were found on patches of flat limestone reef among seagrass. The third site, "Sand Gutters," also at 6-m depth, was established among parallel gutters of limestone reef in between raised ridges of sand. The dominant seagrasses at each site were Posidonia spp. but there were small patches of Amphibolus antarctica and a variety of macroalgae. Although the sites varied greatly in habitat, few abalone in each area reached the legal minimum size (130 ram), and their densities were consistently high prior to thinning. Commercial abalone divers reported that these areas were rarely fished because large abalone were always scarce.

For the translocation experiment two sites where abalone grew to large sizes were established approximately 1 km from the "stunted" sites. Both these sites contained abalone at densities and sizes typical of productive commercial fishing grounds, with an average size approximately 15 mm larger than the slow growth sites. Commercial divers had not fished at either fast growth site for at least 3 y. These sites were surrounded by luxuriant stands of Posidonia spp. and Amphibolus antarctica, with a large loading of epiphytic algae (suitable greenlip food) during surveys in March and September.

Abalone were removed from each fast growth site and a proportion was tagged on the boat and returned to the bottom as fast growth controls. The remainder were removed from the site to allow the translocation of tagged abalone from slow growth sites while maintaining the original density of abalone at the site. Densities of abalone at the two fast growth sites approximated those of slow growth sites. Abalone removed during thinning of the areas at the Lighthouse and Aggregation sites were tagged at the surface and translocated to Fast Growth Sites 1 and 2 respectively. Surface exposure times were similar for all treatments. At the fast growth sites similar numbers of treatments and control abalone were mixed within the same habitat. A second control area was established at each of the Lighthouse and Aggregation sites. Abalone were removed from the bottom and tagged at the surface before being returned, to replicate the method of tagging for transplants and at the fast growth sites.

Abalone were recaptured and measured from each area in September. Incremental growth was standardized to 180 days and regressed against the release length. Because growth increment is expected to vary with initial size, size was included as a covariate, and size-specific responses to treatments were expected. Because Shepherd and Hearn (1983) found growth differences between sexes at one of the sites they investigated, initial ANCOVAs for each site and treatment were set up to test whether growth varied between the sexes. Because we did not detect significant differences between sexes (P > 0.05), the data were pooled for subsequent analyses. Wells and Mulvay (1995) also found no significant differences in growth rate between sexes for Haliotis laevigata in Western Australia.

ANCOVA assumes the covariate is similarly distributed between treatments (Quinn & Keough 2002). In the thinning experiment only data from the Aggregation site contained a different range of sizes for the two treatments. Trimming the data to contain only individuals greater than 90 mm made no difference to the significance of the test and therefore the data from the thinning experiment were not trimmed. In the analysis of the translocation experiment the data were trimmed to include only abalone between 80 mm and 160 mm, to ensure all treatments extended over the same range of sizes. We checked whether the relation of growth increment to initial length was linear by plotting the mean increments of 10-mm size classes. Confidence intervals were calculated for asymptotic lengths (the X axis intercept) from each regression using formulae for X axis intervals in Snedecor and Cochran (1967).

RESULTS

At the Aggregation and Sand Gutters sites the densities remained similar over the 6-mo period: at the Aggregation site density was reduced from 33% of original density after thinning to 29% in September, and at the Sand Gutters site it increased slightly from 33% to 36% (Table 1). The Lighthouse site showed an increase in density consistent with some degree of immigration, from 33% after thinning to 65% after 6 mo (Table 1). The proportion of tagged to untagged individuals reduced from 100% to 19% 6 mo later, and only 37% of all tagged abalone were recaptured within the area. Nevertheless, the average density (the mean of 33% after thinning and 65% after 6 mo) is approximately half of the original abundance. Control area densities remained about the same at the Aggregation site but reduced slightly at the Lighthouse and Sand Gutters sites. At these two sites the average decrease in density was only 10% after 6 too.

Analysis of covariance was performed using data from each of the three sites. The treatment by length interactions were significant; that is the slope of growth rate versus length differed between the control and thinned areas, at each site (Table 2). These tests produce conclusions that relate to each site and might be due to chance differences between the two areas. Sites were then used as replicates to test the effects of density reduction. The treatment by length interaction was tested against the 3-factor interaction because sites were a random factor. This test was significant (F = 67.969 df = 1, 2; P < 0.05), and we conclude that thinning changes the relation between size and growth.

At all sites smaller abalone showed more rapid shell growth in response to thinning (Fig. 2). The size-specific rates of growth at the Aggregation and Sand Gutter sites were similar when compared between control areas and between thinned areas. At the Lighthouse site however, growth rates were much higher, such that size-specific growth at the Lighthouse control was similar to growth in the thinned areas at the other two sites. Nevertheless, the magnitude of the difference between treatments and controls at any given length was very similar at all three sites. The asymptotic lengths of the abalone (X axis intercept) were similar at all sites and treatments, except for the controls at the Aggregation site (Table 3).

[FIGURE 2 OMITTED]

The translocation experiment data were analyzed separately for each site as fast growth control versus translocated and slow growth control versus translocated (Table 4). In analyses of covariance for the translocated versus slow growth controls the treatment by length interaction was highly significant at the Aggregation site (Table 4) but not at the Lighthouse site, probably because there were only 46 recaptures at the Lighthouse slow growth control site. There was, however, a highly significant difference between thinning treatments at the Lighthouse site (Table 4). For translocated versus fast growth controls there was no significant treatment by length interactions, so this interaction term was omitted from the model. In both cases there were highly significant differences between treatments (Table 4). No analysis using sites as replicates was performed because there were only two sites in the experiment.

For translocated abalone from both the Lighthouse and Aggregation sites, the patterns of growth were consistent (Fig. 3). Translocated abalone responded in a similar manner to thinned abalone when compared with slow growth controls, with smaller abalone showing more rapid growth in response to the change of habitat, whereas there were no significant differences in asymptotic length (Table 5). When compared with fast growth controls, translocated abalone grew at a consistently slower rate across all lengths (Fig. 3), resulting in significant differences in asymptotic length at both sites (Table 5). Once again the magnitude of the response in both cases was similar, despite differences in growth rates between the sites.

[FIGURE 3 OMITTED]

DISCUSSION

The growth differences between thinned and control areas were consistent at all three sites despite differences in the magnitude of the density reduction. There were indications of substantial immigration leading to increasing density within thinned areas after 6 mo at the Lighthouse, but not at other sites. There was minimal cryptic habitat and therefore emergence was not likely to have contributed to the increase in density. Presumably, abalone migrated into the area because of the relatively greater abundance of food. At both the Lighthouse and Sand Gutters sites a small reduction in the abalone density within control areas, only 10 m away, was observed. Officer et al. (2001) obtained identical patterns of movement and reaggregation during density reduction experiments of blacklip abalone, and they suggested that the reduction in density within control areas was an indirect consequence of redistribution of individuals into the thinned areas. Be cause of the closeness between control and treatment areas in our experiment, decreases in control density may also have been a consequence of movement into thinned areas, but we did not find any tagged controls in thinned areas.

We expect that tag loss was minimal over our 6-mo experiment, but note that because we estimated final density from the number tagged and the proportion of tagged to untagged during the recapture dives, tag loss would not have affected our density estimates. Mortality of adult abalone over 6 mo would be expected to be very low. The average density over 6 mo at the Lighthouse site was 50% of the original density in the thinned areas and 90% of original density in control areas, effectively equating to a 40% reduction. The average density reduction at the Aggregation and Lighthouse sites were 70% and 60% respectively. Despite the extent of movements at the Lighthouse site, an average density reduction of only 40% was sufficient to establish significant differences in growth rate between thinned and control areas, comparable with the other two sites.

The consistent differences observed led to the overall test result: that density reduction led to an increased slope of the increment versus length regression. The slope of these regressions reflects the Von Bertalanffy growth parameter k, and the X intercept approximates L[infinity] (Day & Fleming 1992). It is clear that for slow growing populations of greenlip abalone, density reduction increases k, but in general it seems not to change L[infinity] much, at least in the short term. This response may reflect the preference of juvenile abalone to direct extra resources into shell growth, whereas larger, older individuals may be placing more of these resources into maintenance and reproduction, a trade-off common to many species (Stearns 1992). L[infinity] was significantly greater in the control area at the Aggregation site than at all other controls and treatments. This difference may be a consequence of the relatively small number of large abalone in the thinned treatment at this site.

At the fast growth sites, a greater food supply was clearly evident during both March and September, because areas of adjacent seagrass had much larger epiphyte loads compared with slow growth sites. These differences in food availability presumably led to the differences in growth and asymptotic length between slow growth and fast growth controls.

We hypothesized that individuals exposed to the same food supply should respond similarly in regard to growth regardless of their growth history. Abalone translocated from slow growth to fast growth sites obtained the same k as the fast growth control population, reflected in the parallel lines of regression at both sites, but grew at a consistently slower rate for all sizes and so that there was a significantly smaller L[infinity]. This suggests that the growth history of an individual does not affect its future k but will affect its maximum size. This difference in average L[infinity] might be expected if there is an age effect on growth, given that slow growing individuals at any given size are likely to be older than fast growing abalone of the same size. An alternative hypothesis is that the previous history of abalone determines their resource allocation, so that the larger transplanted abalone may devote almost all of the resources they have available beyond maintenance requirements, to reproduction. It would be useful to study changes in growth and fecundity over longer periods to see if this pattern persists.

Direct comparison of translocated abalone with the thinning experiment controls could not be made because the latter were tagged in situ. Translocated abalone had to be handled on the boat so all controls were treated in the same way. When the growth of translocated abalone was compared with slow growth controls, the pattern of growth response was the same as for thinning and was similar among sites. On the basis of these results it is reasonable to speculate that the manipulation of density reduction and change of habitat have increased the quantity of food available to the individuals within that population, and this led to the size-specific growth response seen in both experiments over the short periods of this study.

The magnitude of the response in shell growth to the density reductions of 40% to 70% in unfished populations is critical for stock assessment. Many biomass dynamic fishery models have adopted sustainability criteria such as failure to maintain a minimum proportion of original biomass, to trigger management intervention. For these triggers the values used are usually around 40%, equivalent to a loss of 60% of the original unfished biomass (Smith & Smith 2002). For abalone, Shepherd & Baker (1998) suggested that a minimum egg production of 40% to 50% be maintained. Depending on size limits, these values are likely to equate to density reductions within the range of this study. Thus it is important that assessments of areas are not carried out soon after a small area is fished. The magnitude of the density dependent growth response we observed, despite the short-term nature of our experiment, suggests that biomass reduction of legal sized abalone by fishing would be at least partly compensated by growth over longer periods.

In the longer term, density dependent growth would accelerate stock declines in fisheries with a legal minimum length (LML), because fishing would increase the growth rate of sublegal sized abalone, rendering them vulnerable to fishing after a shorter period of time. Density dependent growth becomes compensatory however, when the increase in size reduces mortality or increases fecundity (Rose et al. 2001). For H. laevigata there is clear evidence of both size dependent mortality (Shepherd & Breen 1992) and rapidly increasing fecundity with increasing size (Shepherd et al. 1992). Therefore the significant increases in growth rate that we observed could be expected to act as a regulatory mechanism that promotes increased population growth to compensate for losses caused by density reduction.

McAvaney et al. (this volume) have shown that juveniles at low density will reach larger sizes and mature earlier than juveniles at high density. Their study involves the manipulation of densities of juvenile greenlip abalone (<90 mm) outplanted into the wild. It is important to determine whether a reduced density of adult sizes leads to faster growth of juveniles that occupy more cryptic habitats and thus might be more affected by juvenile than by adult density

In our study we have shown that compensatory growth will occur in fished abalone populations. We hypothesize that larger abalone were allocating extra resources into reproduction and that if this were the case, further compensatory effects would occur. If fishing large adults leads not only to faster growth of small adults but also to faster recruitment of adults from the juvenile size classes, then these mechanisms in combination may assist to stabilize adult stocks against the impact of fishing and at least partly maintain the egg production capacity of abalone populations, which will in turn increase prospects for their sustainability.

TABLE 1.

Estimation of abundance from the proportion of tagged to untagged
individuals encountered within sites at Tiparra reef.

                               March      Number     % Abundance
    Site        Treatment    Abundance    Tagged    Post-thinning

Lighthouse       Thinned        603        201           33%
                 Control        246        246          100%
Aggregation      Thinned        810        270           33%
                 Control        269        269          100%
Sand Gutters     Thinned        354        118           33%
                 Control        135        135          100%

                                               % in        Estimated
                             Total Tags    Aggregations    September
    Site        Treatment    Recaptured     with Tags      Abundance

Lighthouse       Thinned         74            19%            389
                 Control         99            51%            194
Aggregation      Thinned        142            61%            233
                 Control        215            81%            265
Sand Gutters     Thinned         85            67%            127
                 Control         97            93%            104

                             % of March
    Site        Treatment    Abundance

Lighthouse       Thinned        65%
                 Control        79%
Aggregation      Thinned        29%
                 Control        99%
Sand Gutters     Thinned        36%
                 Control        77%

TABLE 2.

Tests of differences between slopes in thinned versus control areas,
using analysis of covariance at each site.

                                             Mean       F-        P-
   Site              Source          DF     Squares    ratio    values

Aggregation    Treatment x Length      1    55.891     7.945    0.007
               Error                 351     7.457
Gutters        Treatment x Length      1    36.681     6.032    0.015
               Error                 175     6.081
Lighthouse     Treatment x Length      1    73.208     8.227    0.005
               Error                 169     8.899

TABLE 3.

Estimated asymptotic lengths and confidence intervals based on X
axis intercepts of regressions for control and treatments at each site
of the density reduction experiment.

    Site        Treatment     n     -95%CL    L[infinity]    +95%CL

Aggregation      Control     213    149.2       153.9        158.7
                 Thinned     142    135.5       142.1        148.6
Lighthouse       Control      99    137.8       144.1        150.5
                 Thinned      74    137.6       143.3        149.1
Sand Gutters     Control      96    137.9       141.3        144.8
                 Thinned      85    139.3       145.9        152.5

TABLE 4.

ANCOVA outputs for tagged abalone translocated from the Lighthouse and
Aggregation sites compared to slow growth and fast growth controls.

Translocated from    Comparison       Source      DF       MS

Aggregation         Slow V Trans.   Interaction     1    126.743
                                    Error         292      6.450
Lighthouse          Slow V Trans.   Interaction     1     12.044
                                    Error         124      5.531
Lighthouse          Slow V Trans.   Treatment       1    107.655
                                    Error         125      5.584
Aggregation         Fast V Trans.   Treatment       1   1609.648
                                    Error         299      8.060
Lighthouse          Fast V Trans.   Treatment       1    699.589
                                    Error         222      7.179

Translocated from    Comparison       Source      F-ratio   P-value

Aggregation         Slow V Trans.   Interaction    19.650    <0.001
                                    Error
Lighthouse          Slow V Trans.   Interaction     2.177     0.143
                                    Error
Lighthouse          Slow V Trans.   Treatment      19.281    <0.001
                                    Error
Aggregation         Fast V Trans.   Treatment     199.701    <0.001
                                    Error
Lighthouse          Fast V Trans.   Treatment      97.456    <0.001
                                    Error

TABLE 5.

Estimated asymptotic lengths and confidence intervals based on
regressions of control and treatment data at each site of the
translocation experiment.

   Site         Treatment       n     -95%CL    L[infinity]    +95%CL

Aggregation    Fast control    174    151.1        156.5       162.0
               Translocated    135    133.6        139.7       145.9
               Slow control    165    133.4        138.1       143.7
Lighthouse     Fast control    141    153.4        158.9       164.4
               Translocated     84    141.1        146.8       152.5
               Slow control    461    133.4        137.6       141.7

ACKNOWLEDGMENTS

The authors thank the Central Zone divers and deckhands of South Australia for their contribution in the field; Michael Tokley and Bob Pennington for administrative assistance; Sylvain Huchette for the very valuable assistance in experimental design and lead role in fieldwork. The authors also thank Harry Gorfine and Matt Reardon for constructive criticism of the manuscript; Scoresby Shepherd, James Brook, Simon Hart, Matt Reardon, Thor Saunders, Steven Mayfield, and Brian Davies for their assistance in field work. This research was conducted as a component of a project established between the University of Melbourne and the Abalone Industry Association of South Australia, funded under the Australian Research Council (ARC) Strategic Partnerships with Industry Research and Training (SPIRT) Scheme

LITERATURE CITED

Babcock, R. & J. Keesing. 1999. Fertilization biology of the abalone Haliotis laevigata: Laboratory and field studies. Can. J. Fish. Aquat. Sci. 56:1668-1678.

Barki, A., N. Gur & I. Karplus. 2001. Management of interspecific food competition in fish-crayfish communal culture: the effects of the spatial and temporal separation of feed. Aquaculture. 201:343-354.

Brazeiro, A. & O. Defeo. 1999. Effects of harvesting and density-dependence on the demography of sandy beach populations: the yellow clam Mesodesma mactroides of Uruguay. Mar. Ecol. Prog. Ser. 182:127-135.

Capinpin Jr, E. C., J. D. Toledo, V. C. Encena Ii & M. Doi. 1999. Density dependent growth of the tropical abalone Haliotis asinina in cage culture. Aquaculture. 171:227-235.

Day, R. W. & A. E. Fleming. 1992. 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, UK: Blackwell. pp. 141-168.

Emmett, B. & G. S. Jamieson. 1989. An experimental transplant of northern abalone, Haliotis kamtschatkana, in Barkley Sound, British Columbia. Fish. Bull. 87:95-104.

Huchette, S. M. H., C. S. Koh & R. W. Day. 2003. Growth of juvenile blacklip abalone (Haliotis rubra) in aquaculture tanks: effects of density and ammonia. Aquaculture. 219:457-470.

Hunt, J. W., M. S. Foster, J. W. Nybakken, R. J. Larson & E. E. Ebert. 1995. Interactive effects of polyculture, feeding rate, and stocking density on growth of juvenile shellfish. J. Shellfish Res. 14:191-197.

Kamermans, P., H. W. van der Veer, L. Karczmarski & G. W. Doeglas. 1992. Competition in deposit- and suspension-feeding bivalves: experiments in controlled outdoor environments. J. Exp. Mar. Biol. Ecol. 162:113-135.

Marshall, P. A. & M. J. Keough. 1994. Asymmetry in intraspecific competition in the limpet Cellana tramoserica (Sowerby). J. Exp. Mar. Biol. Ecol. 177:121-138.

McAveney, L. A., R. W. Day, C. D. Dixon & S. M. Huchette. (this volume) Gonad development in seeded Haliotis laevigata: growth environment determines initial reproductive investment. J. Shellfish Res. Under review.

McShane, P. E. 1995. Estimating the abundance of abalone: the importance of patch size. Mar. Freshwater Res. 46:657-662.

McShane. P. E. & J. R. Naylor. 1995a. Density independent growth of Haliotis iris Martyn (Mollusca: Gastropoda). J. Exp. Mar. Biol. Ecol. 190:51-60.

McShane, P. E. & J. R. Naylor. 1995b. Small-scale spatial variation in growth, size at maturity, and yield- and egg-per-recruit relations in the New Zealand abalone Haliotis iris. N. Z. J. Mar. Freshwater Res. 29:603-612.

McShane, P. E., D. R. Schiel, S. F. Mercer & T. Murray. 1994. Morphometric variation in Haliotis iris (Mollusca: Gastropoda): analysis of 61 populations, N. Z. J. Mar. Freshwater Res. 28:357-364.

Mgaya, Y. D. & J. P. Mercer. 1995. The effects of size grading and stocking density on growth performance of juvenile abalone, Haliotis tuberculata Linnaeus. Aquaculture 136:3-4.

Officer, R. A., C. D. Dixon & H. K. Gorfine. 2001. Movement and reaggregation of the blacklip abalone, Haliotis rubra leach, after fishing. J. Shellfish Res. 20:771-779.

Prince, J. D. 1991. A new technique for tagging abalone. Aust. J. Mar. Freshwater Res. 42:101-106.

Quinn, G. P. & M. J. Keough. 2002. Experimental design and data analysis for biologists. Cambridge: Cambridge University Press. 537 pp.

Rose, K. A., J. H. Cowan, Jr., K. O. Winemiller, R. A. Myers & R. Hilborn. 2001. Compensatory density dependence in fish populations: importance, controversy, understanding and prognosis. Fish and Fisheries 2:293-327.

Shepherd, S. A. 1973. Studies on southern Australian abalone (genus Haliotis). I. Ecology of five sympatric species. Aust. J. Mar Freshwater Res. 24:217-257.

Shepherd, S. A. 1986. Studies on southern Australian abalone (genus Haliotis). VII Aggregative behaviour of 14. laevigata in relation to spawning. Mar. Biol. 90:231-236.

Shepherd, S. A. & J. L. Baker. 1998. Biological reference points in an abalone (Haliotis laevigata) fishery. Can. Spec. Publ. Fish. Aquat. Sci. 125:235-245.

Shepherd, S. A. & P. A. Breen. 1992. Mortality in abalone: its estimation, variability and causes. In: S. A. Shepherd, M. J. Tegner & S. A. Guzman del Proo, editors. Abalone of the world: biology, fisheries, and culture. Oxford, UK: Blackwell. pp. 276-304.

Shepherd, S. A., C. Godoy & S. M. Clarke. 1992. Studies on southern Australian abalone (genus Haliotis). XV. Fecundity of H. laevigata. J. Malac. Soc. Aust. 13:115-121.

Shepherd, S. A. & W. S. Hearn. 1983. Studies on Southern Australian abalone (genus Haliotis). IV. Growth of H. laevigata and H. ruber. Aust. J. Mar Freshwater Res. 34:461-475.

Smith, A. D. M. & D. C. Smith. 2002. Towards an improved basis for setting TACs in the South East Fishery. Report to AFMA, 44. Canberra: Department of Environment.

Snedecor, G. W. & W. G. Cochran. 1967. Statistical methods. Iowa: Iowa State University Press. 593 pp.

Stearns, S. C. 1992. The evolution of life histories. Oxford: Oxford University Press. 249 pp.

Stoner, A. W. 1989. Density-dependent growth and grazing effects of juvenile queen conch Strombus gigas L. in a tropical seagrass meadow. J. Exp. Mar. Biol. Ecol. 130:119-133.

Talman, S. G. & M. J. Keough. 2001. Impact of an exotic clam, Corbula gibba, on the commercial scallop Pecten fumatus in Port Phillip Bay, South East Australia: evidence of resource restricted growth in a subtidal environment. Mar. Ecol. Prog. Ser. 221:135-143.

Vromant, N., C. Q. Nam & F. Ollevier. 2002. Growth performance of Barbodes gonionotus (Bleeker) in intensively cultivated rice fields. Aquaculture 212:167-178.

Wells, F. E. & P. Mulvay. 1995. Good and bad fishing areas for Haliotis laevigata: A comparison of population parameters. Mar. Freshwater Res. 46:591-598.

Zoology Department, The University of Melbourne, Victoria