Feeding preferences of the abalone Haliotis iris in relation to macroalgal species, attachment, accessibility and water movement.

ABSTRACT Haliotis iris is a species of abalone common on rocky reefs in southern and central New Zealand. This study examined the poorly understood feeding habits and preferences of H. iris in a series of laboratory experiments. Generally, H. iris consumed the blades of brown algae over red and green algae. However, when upright whole plants were given to H. iris, the highly preferred kelp Lessonia variegata was consumed in lower proportions than the less preferred but more accessible red alga Gigartina circumcincta. H. iris were less capable of reaching the blades or consuming the stipe of L. variegata, which has a stipe of ~100-350 mm high. H. iris consumed greater amounts of drift over benthic L. vareigata. Water movement appeared to inhibit the active grazing of H. iris, but not the drift-trapping behavior, resulting in lower overall feeding rates for abalone under conditions of higher water movement. Abalone consumed fresh and aged algae equally. We conclude that H. iris feeds primarily on drift algae because preferred food sources are more accessible as drift than as attached macroalgae, and because this may be a more successful foraging strategy in the high flow environment this species commonly inhabits.

KEY WORDS: feeding preferences, grazing, Haliotis iris, drift algae, algal accessibility, plant-herbivore interactions, water motion

INTRODUCTION

Herbivore food choice can play a strong role in determining algal distribution and diversity (Lubchenco 1978). Hence, it is important to understand the dietary preferences and foraging behavior of locally abundant herbivores in marine systems dominated by macroalgae. Abalone are large molluscan grazers that can be locally abundant on temperate rocky reefs throughout the world (Geiger 2000), yet their role in influencing the macroalgal communities they inhabit is poorly understood and maybe more complex than previously hypothesized (see Tegner & Dayton 2000). Abalone feeding preferences have been well studied in many regions, particularly through the use of choice experiments. It is believed that abalone from the northern hemisphere (Guzman del Proo et al. 2003, Alcantara & Noro 2005) and South Africa (Wood & Buxton 1996, Day & Branch 2002) prefer to feed on species of brown algae, whereas those from the south-east Asia (Tahil & Juinio Menez 1999) and Australia (Shepherd & Steinberg 1992, McShane et al. 1994) prefer to feed on species of red algae.

In New Zealand, there have been conflicting reports on the feeding preferences of the abalone Haliotis iris (commonly referred to as the blackfoot abalone of paua). Some studies have reported that this species feeds preferentially on species of red algae, similar to Australian abalone (Poore 1972, Marsden & Williams 1996). Other studies have shown this species feeds primarily on brown algae over reds and greens (Sinclair 1963, Tunbridge 1967, Dutton & Tong 1981). However, a more recent paper has demonstrated that H. iris has no preferences amongst these groups of algae at all (Taylor & Steinberg 2005). Factors such as chemical defense (Shepherd & Steinberg 1992) and morphology (McShane et al. 1994) are believed to play a large role in determining abalone feeding preferences, though they remain largely untested for H. iris.

Any potential effects of abalone on algal communities are likely to be strongly related to foraging behavior. Abalone forage by actively grazing attached benthic algae, by sedentary capture of passing drift algae, or by browsing the substrate for animal and plant matter (Shepherd 1973, Miner et al. 2006). Active grazing on live algae has been recorded less frequently, but may have a more direct effect on algal communities than feeding on detached, decomposing material. H. iris is believed to feed predominately on drift algae (Poore 1972), though it is often associated with areas of reduced macroalgal cover (Shears 2007, Shears & Babcock 2007), indicating this species may also actively graze.

Abalone may prefer to feed on drift algae because degradation is believed to reduce the chemical and morphological defences of macroalgae (Duggins & Eckman 1997, Rothausler & Thiel 2006). Drift feeding may also be an adaptive response to an underexploited food source as implied by Estes et al. (2005), although abalone foraging behavior could also be plastic in response to the availability of drift algae. The availability of drift algae has been hypothesized to increase with increasing water flow in a given habitat (Shepherd 1973), suggesting that drift feeding may be an effective strategy in areas of high water movement. In support of this hypothesis, several species of abalone have higher growth rates in more exposed habitats than in sheltered habitats (Sainsbury 1982, McShane et al. 1988, Keesing & Wells 1989).

Brown algae are generally more abundant in the areas of high-water motion preferred by many species of abalone (Choat & Schiel 1982), but abalone may have limited access to the more edible blades of brown algae in the field, because of the typically large, tall stipes characteristic of many species in this division. The height of the blades above the substrate may provide a refuge from grazing by abalone, and drift-feeding may be a mechanism that allows abalone to gain access to a food-source that is otherwise largely inaccessible. To date however, apart from qualitative observations on foraging behavior in the field (e.g., Shepherd 1973, Tutschulte & Connell 1988), no published study has attempted to quantify the drift-feeding behavior of any abalone species, of explicitly examine attributes of drift and benthic algae and how they may influence abalone foraging behavior.

There were two general purposes of this study; the first was to examine feeding preferences by H. iris by testing the following hypotheses: (1) H. iris prefers to feed on species of brown algae over species of reds and greens and (2) H. iris prefers to feed on older, more degraded algae over fresh algae. The second was to examine the role of mechanistic factors, such as accessibility and water movement, on feeding by testing three additional hypotheses: (1) H. iris feeds more on unattached algae than attached algae; (2) Any preference for unattached algae will increase with increasing water movement; and (3) H. iris feeds more on easily accessible algae (i.e., blades that are lower in height).

MATERIALS AND METHODS

Testing of these hypotheses were conducted in a controlled laboratory environment, because field observations provided little information on abalone feeding (i.e., after 3 h of night and 12 h of day observations only 11 H. iris individuals were seen feeding). All experiments were conducted at Victoria University Coastal Ecology Laboratory (VUCEL), Wellington, New Zealand. All adult H. iris (shell length 110-134 mm) were collected in the vicinity of Wellington on the south coast of the North Island, New Zealand (41[degrees]20.9'S, 174[degrees]45.9'E) and kept in aquaria with flow-through seawater (13[degrees]C [+ or -] 1.5[degrees]C) on a 12 h light: dark regime, unless stated otherwise. Abalone were conditioned for one month prior to each experiment, by feeding them the same species of algae subsequently used in the experiment, to remove any biases based on previous diets in the field (after Cronin & Hay 1996). For experiments using multiple species of algae, every species that was used was made available to the abalone during conditioning. All algae used in this study were collected from the Wellington region and used on the day of their collection unless otherwise stated.

Pair-wise Feeding Assays

To examine H. iris feeding preferences for different species of algae, sixty adult abalone (104-134 mm) were placed into individual mesh containers (300 x 350 x 70 mm). The mesh (2 x 2 mm apertures) allowed water flow but stopped large algal detritus from flowing out, and was re-enforced by plastic strips along the sides and bottom. Control and treatment containers were placed into the same large water bath (1,600 x 5,600 x 100 mm), so that control and treatment algae experienced similar conditions (see Prince et al. 2004 for a discussion of why this is necessary). One gram (blotted wet weight to the nearest 0.01 g) of blade tissue from each of the two algal species used in the particular pair-wise combination was placed into each treatment and control. After 24 h, the amount of algae remaining was removed and weighed to the nearest 0.01 g (see data analysis section later). The algae used were the browns Cystophora retroflexa, Ecklonia radiata, Lessonia variegata, Zonaria turneriana, Carpophyllum maschalocarpum; the reds Gigartina circumcincta, Corallina officinalis, Pterocladia lucida, Asparagopsis armata, and Rhodophyllis gunnii; the greens: Caulerpa brownii, Caulerpa geminata, Caulerpa flexilis, and Ulva sp. These species of macroalgae were used because they were believed to be the species most available to H. iris in the region (C. Cornwall personal observation, Shears & Babcock 2007). Only four species of green algae were used because they were the only green species readily available in the subtidal habitat H. iris occupies in the study location. Algae were used within three days of collection.

The experimental design was replicated such that every possible pair-wise combination of the 14 species of algae (91 pairs) was replicated at least 6 times (i.e., using 6 different individual abalone for each combination, allocated using random numbers), whereas ensuring that any one pair-wise combination was not red to the same individual (60 H. iris were used) more than once. This assay was conducted between September and October 2006.

Algal Age

This experiment consisted of two factors: algal species (Lessonia variegata and Gigartina circumcincta) and algal age (fresh or 7 days old). L. variegata and G. circumcincta were chosen because of their similar blade thickness and toughness, and their high local abundance (C. Cornwall, personal observation). Older algae that were "aged" were collected a week prior to the experiment, and held in a tank (600 x 600 x 300 mm) with flowing seawater at 12.5[degrees]C [+ or -] 0.5[degrees]C (similar to the 5 days of aging used by Marsden & Williams 1996). Algae were softer and more decomposed than fresh algae when assessed visually after 7 days.

Experimental tanks (300 x 600 x 300 mm) were supplied with unfiltered flowing seawater. A plastic mesh (10 x 10 mm aperture) partition was placed in the middle of each tank, to divide the tank into two sections: a treatment section where 3 abalone were placed, and a control section where only algae were placed (as recommended by Prince et al. 2004). Control algae were needed to account for autogenic changes in mass. Fifty grams of blade tissue of each type of algae were each weighted to the bottom of the tank with a peg.

The entire experiment tan for 70 days starting October 2005, with algae being replaced and reweighed every 7 days (by this time 14-day-old algae were substantially more degraded than 7-day-old algae). The experiment was conducted for this length of time to examine feeding activity over time, because it is possible that the results of shorter experiments may be caused by spikes in feeding activity. There was minimal mortality of abalone in the experiment, and those replicates were excluded from statistical analyses.

Algal Attachment and Water Motion

For this experiment, only L. variegata was used and the factors examined were attachment (4 levels) and water motion (added with a pump, or not). The tank set up was the same as in the previous experiment. L. variegata blades were cut into 30-g pieces and attached by pegs to the mesh partition, at three different points: high (250 mm from the bottom of the tank), medium (125 mm from the bottom of the tank), and low (bottom of the tank). This simulated situations in the field, where blades of different species of macroalgae are located at different heights above the substrate. Abalone could climb up the sides of the tank though, mimicking the natural settings where plants grow on rocks at different angles, but are more accessible to benthic grazers at lower heights. Unattached algae (30 g of blade tissue) were also placed in the tank and left as drift. There were 20 replicate tanks with all four treatments in each tank simultaneously. Half of the replicates of each treatment had additional water motion provided by a Water-werks 350 1 [h.sup.-1] pump (average velocity ~26.1 [+ or -] 5.7 cm [s.sup.-l]), whereas in the remaining tanks, low flow water movement was provided by the running seawater system (2.1 [+ or -] 0.4 cm [s.sup.-l]). Velocity was initially determined by recording how long it takes the object to move a certain distance, then confirmed a posteriori using an Acoustic Doppler Velocimeter. Water velocity in the tanks containing pumps varied between 15-35 cm [s.sup.-1] depending on the proximity to the pump. As such, the velocity given is an average velocity in the area where macroalgae were placed. The pumps simulated some subtidal conditions where a unidirectional current runs through kelp beds, and were approximate to conditions of high water velocity for the subtidal environment (Hurd 2000). Drift algae in treatments with water pumps added were constantly moving in turbulent motion, whereas those in the low flow treatments tended not to move.

Three abalone were placed into the treatment section of each tank, with no abalone on the control side. The entire experiment lasted 70 days starting April 2006, with algae replaced and reweighed every 7 days.

Algal Species and Accessibility

The species L. variegata and G. circumcincta were used in this experiment because L. variegata has a tall stipe--plants can grow to 700 mm above the substrate (C. Cornwall, personal observation)--and was suspected to be a highly preferable food for H. iris (see results section). G. circumcincta, by contrast, was believed to be a less preferable red alga but one that is more accessible because its blades grow close to the substrate. On collection, L. vareigata holdfasts were pulled off the substrate whole, and care was taken not to damage them.

Five large tanks (circular: 1.5 m diameter, 0.4 m high) were placed in a large outside pool (7 m diameter, 1.5 m high) filled with flowing seawater. Each of the three treatment tanks had a plasticized wire mesh cover so no abalone could escape. These tanks were used to better approximate natural settings, because no abalone could access algae from the side in this experiment because of the large size of the tanks. The two remaining tanks were controls (algae only, no abalone).

Four different types of algae (approximately 100 g of each) were placed in each tank: whole G. circumcincta plants (pegged to weights), L. variegata blades only (pegged to weights), G. circumcincta blades pegged to L. variegata stipes, and whole L. variegata plants. Therefore, blades of both species were made directly accessible to benthic abalone on the floor of the tank and also higher in the water column, atop a L. variegata holdfast and stipe (150-350 mm above the substrate).

Twenty abalone were placed into each of the treatment tanks; the entire experiment was run for 28 days starting November 2006, with algae replaced and reweighed every 7 days.

Data Analysis

The amount of algae (in wet weight) eaten per abalone was calculated by using the formula F = [F.sub.l] ([F.sub.2] [+ or -] C), from Tahil and Junio Menez (1999). Where: F = actual food consumption in each given treatment, [F.sub.1] = amount of food offered, [F.sub.2] = remaining amount of food at the end of 24 h, and C = change in control algae without abalone. The change in weight of treatment and control algae were determined at the same time, and control algae were paired with treatment algae to determine F for that particular trial, except for the algal species and accessibility assays (as per Peterson & Renaud 1989). Because the initial weight of algae differed slightly among the treatments in the algal species and accessibility assay, a proportional amount of algae consumed was calculated instead of the total amount. The formula used to determine the proportion of algae eaten was a modified version of the formula used previously: P = F / [F.sub.1]. Where: P = the proportion of algae eaten for that treatment, and F = ([F.sub.1] - [F.sub.2]) x C, [F.sub.1] = initial weight of treatment algae, [F.sub.2] = end weight of treatment algae, C = the average of [C.sub.2] (end weight of control algae)/[C.sub.1] (initial weight of control algae) for that week.

For the pair-wise feeding assays F, actual food consumption, was analyzed with a series of paired t-tests (as recommended by Peterson & Renaud 1989). For all other experiments F values (or P values for the algal species and accessibility experiment) were analyzed using an analysis of variance with repeated measures (ANOVAR). ANOVAR was used as it takes into account the repeated sampling of the same replicate (in this case the same abalone, of group of abalone) over time, whereas an ANOVA does not (for a discussion see Keselman et al. 2001). Greenhouse-Geiser adjusted degrees of freedom were used, as they are the most conservative values. All data met the sphericity assumption using Mauchly's Test of Sphericity. Posthoc Tukey tests were ran when appropriate. Homoscedasticity of variance was tested using Cochran's test (Underwood 1981). All statistics were done in SPSS version 12.

RESULTS

Pair-wise Feeding Assays

Out of the 91 total pair-wise combinations, significant differences (P [less than or equal to] 0.05) in the amount of algae consumed between the two species occurred in 27 combinations (Table 1). Of the pair-wise combinations that differed significantly, 55.5% were cases where the preferred alga was a brown species paired against a red or green species (Table 1, Fig. 1). Of the brown algae that were preferred, 5 cases were L. variegata, 4 were Cy. retroflexa, 3 were Z. turneriana, 2 were Car. maschalocarpum, and 1 was E. radiata. In cases where a species of brown algae was paired with another species of brown algae, both species were consumed in similar amounts. When a species of brown algae was paired against a species of red or green algae, the red or green species was never preferred.

Twenty-six percent of significant pair-wise combinations resulted in a species of red algae being consumed preferentially against another species of red or green algae; in 3 cases the preferred red species was A. armata, in 2 cases Co. officialis, and in 1 case each G. circimcincta and R. gunnii (both when paired against a species of Caulerpa).

Eighteen and a half percent of pair-wise trials with significant differences occurred when a species of green algae was eaten in greater amounts than another species of green or red (in 3 cases the species was Ulva sp., and in 2 it was C. flexilis).

Finally, in 74% of significant pair-wise combinations the species least preferred was one of 4 species: G. circumcincta (22%), C. brownii (18.5%), P. lucida (18.5%) or C. flexilis (15%).

Algal Age

There was no effect of algal age on the per capita amount of algae consumed ([F.sub.1, 296] = 0.059, P = 0.824), although algal species was significant ([F.sub.1, 296] = 61.1, P = 0.004). There was no significant interaction between the two factors ([F.sub.1, 296] = 1.38, P = 0.325). Similar to the pair-wise feeding assay, L. variegata was consumed in greater amounts than G. circumcincta, in this experiment nearly twice as much (Fig. 2). There was also no significant effect of time on grazing rates ([F.sub.9, 296] = 1.134, P = 0.339), nor any interaction between time and species ([F.sub.9, 296] = 1.621, P = 0.109), or between time and algal age ([F.sub.9, 296] = 1.199, P = 0.296).

Algal Attachment and Water Motion Experiment

Water movement ([F.sub.1, 359] = 6.975, P = 0.019) and attachment type and position ([F.sub.1.739, 359] = 11.219, P = 0.001) significantly affected the amount of L. variegata consumed, with no interaction between these two factors [F.sub.1.739, 359] = 1.233, P = 0.304; Fig. 3). Post-hoc Tukey tests showed that abalone consumed more unattached (drift) algae than algae attached at any height off the substratum at both water velocities. There was no effect of time ([F.sub.3.029, 359] = 2.804, P = 0.051), nor was there an interaction between time and attachment ([F.sub.6.231, 359] = 0.802 P = 0.575), time and water movement ([F.sub.3.029, 359] = 1.577, P = 0.209), or all three factors ([F.sub.6.231, 359] = 1.044, P = 0.403). Contrary to the a priori hypothesis, the effect of extra water movement did not increase the amount of algae consumed, but rather decreased total consumption (Fig. 3). Although the interaction was not significant, it seems that adding water movement tended to reduce the consumption of attached algae more than unattached, drift algae.

Algal Species and Accessibility

Abalone consumed a greater proportion of blades that were directly on the substrate compared with those on top of L. variegata stipes ([F.sub.1, 42] = 51.032, P = 0.019), whereas algal species had no effect on the amount consumed ([F.sub.1, 42] = 0.392, P = 0.595) and there was no interaction between the two factors ([F.sub.1, 42] = 6.696, P = 0.123, Fig. 4). There was no effect of time ([F.sub.1.4, 42] = 1.094, P = 0.421), nor any interaction between time and species ([F.sub.161, 42] = 1.125, P = 0.403), time and location ([F.sub.1.4, 42] = 0.597, P = 0.803), or all three factors ([F.sub.1.713, 42] = 0.137, P = 0.849).

[FIGURE 1 OMITTED]

DISCUSSION

H. iris tended to consume more brown algae than reds or greens, although some exceptions existed (i.e., Ulva sp. and A. armata), supporting the hypothesis that it prefers brown algae over reds and greens. Here, we can draw more accurate conclusions than the research that found species of red algae were preferred over browns (Poore 1972, Marsden & Williams 1996), because of the greater number of species combinations used in this experiment, correct statistical analysis, and standardization of abalone diets prior to the study. Conclusions based on preference tests that use a small number of species may vary depending on the choice of species in the test, highlighting the need for preference experiments to include as many dietary items as possible that are available in the system of interest. It is even possible that the 14 different species of algae used in the test were not enough to make reliable hypotheses regarding H. iris feeding preferences, as many more species are available in the field.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

Abalone sometimes prefer macroalgae with high nitrogen contents (Fleming 1995) and no, or weak, chemical defences (Shepherd & Steinberg 1992, Paul et al. 2006), and it is likely that an interaction between nutrient levels and the presence of chemical defense also play a role in determining the preferences of H. iris. The chemical defences of brown algae from Australasia have been perceived as an adequate deterrent from herbivory (Estes & Steinberg 1988, Shepherd & Steinberg 1992, Steinberg et al. 1995, though see Steinberg & Van Alterna 1992). This has helped explain why some previous studies came to the conclusion that H. iris preferred red algae over brown (Poore 1972, Marsden & Williams 1996). However, it is possible that brown algae in New Zealand are less chemically defended than previous authors have hypothesized. Recently it has been suggested that the prehuman levels of herbivore predators in New Zealand were much higher than formerly believed (Shears & Babcock 2002, 2003), and therefore, herbivory may have been lower than previously thought, which suggests that high concentrations of defensive chemical compounds were not needed for macroalgae in this region. Also, a recent analysis of stable isotopes has discovered that the Australian H. rubra likely feeds predominantly on brown macroalgae in the field (Guest et al. 2008). This, along with the fact that some Australian herbivore predators have also been fished down (Stuart-Smith et al. 2008), indicates that Australian brown macroalgae may also have lower chemical defences than previous believed.

[FIGURE 4 OMITTED]

H. iris consumed more unattached algae over attached algae at any height at high and low velocities, which indicates that attachment state of the algae had a greater influence than its accessibility in determining consumption rates. However, the interaction between water velocity and accessibility was complex. Increased water movement tended to reduce the overall level of grazing on attached algae, rather than increase drift-feeding, contrary to our hypothesis. There was an overall reduction in consumption rates of all three heights of attached algae when water velocity was increased, but this reduction was only significant for low algae. This was probably caused by small amounts (~1 g per capita per week) of all three types of attached algae being consumed haphazardly when abalone were both drift-feeding and actively grazing (Fig. 3.).

This suggests that H. iris foraging may be dependent on environmental conditions, as drift-feeding appeared to always be prevalent, but under conditions of lower flow (when abalone may have a reduced probability of encountering drift), active grazing tended to increase. When drift feeding, abalone raise their shell in a distinctive posture (Shepherd 1973). This posture was more common in treatments with higher water flow, indicating that abalone may be able to identify differences in water velocities and change their foraging behavior to suit the appropriate flow rate. Therefore, if grazing behavior increases under conditions of lower water flow in the field, water velocity may play a relatively important role in determining the ecological effects of H. iris feeding.

In this study, the age of algae did not play a role in the feeding preference of H. iris. Previous studies have found that for amphipods and isopod mesograzers, older algae is preferred and that grazer survival increased when living on older algae, compared with fresh algae (Norderhaug et al. 2003, Rothausler & Thiel 2006). Gastropod mesograzers however, do not have increased survival on older plants (Norderhaug et al. 2003). For example, Rothausler and Thiel (2006) found that only after 12 days of decomposition did Lessonia nigrescens become more palatable to mesograzers when compared with fresh algae. Given that L. variegata was one of the most highly preferred species in this study, it is likely that chemical defences of this species are lower than those of L. nigrescens.

The role of algal toughness and calcification as deterrents have been investigated for a range of marine herbivores (e.g., Littler & Littler 1980, Lubchenco & Gaines 1981, Steneck & Watling 1982 and references therein), including abalone (Shepherd & Steinberg 1992, McShane et al. 1994). Algal whiplash can also defend kelp from sea urchin herbivory (Kawamata 1998, Gagnon et al. 2006). Apart from these mechanical defences, the role that macroalgal morphology plays in determining subtidal herbivore consumption rates has been relatively untested. The current study shows that algal morphology may play a role in determining the intensity of herbivory that macroalgae may encounter in the field. Even though L. variegata was the preferred diet of H. iris when blades were made easily accessible, whole L. variegata was difficult for the abalone to consume because of the height of its blades above the substrate. Taller plants, such as the local Ecklonia radiata, may be further protected than L. variegata from abalone herbivory, because of even larger distances between the blades and the substrate. This form of plant defense will only protect against benthic herbivores, which are unable to climb plants to reach blade material. Other herbivores, such as more mobile sea urchins and mesograzers will not be deterred by these defences during periods of low water movement.

Algal removal by benthic herbivores in the field does not always correlate with laboratory feeding preferences (Schiel 1982). In nature, a number of different factors will influence the rate of algal removal by herbivores. These factors include those investigated in the current study, such as algal accessibility and morphology, the degree of water motion, and the availability of drift algae, an important alternative food source. Other factors that could play a role in natural herbivory rates include the presence or absence of competitors (e.g., Tegner & Levin 1982), predators (Byrnes et al. 2006), facilitators (e.g., Fletcher 1987), and alternative food sources that may be undetectable to the investigators, such as particulate organic material and diatoms (Pearse & Pearse 1973).

These experiments suggest that H. iris prefers to feed primarily by drift-feeding, rather than by actively grazing; a tendency influenced by water flow. The results also suggest that drift algae is consumed by H. iris, not because of the age of the algae, but most likely because of two factors: (1) highly preferred brown algae, such as L. variegata, are more accessible as drift than as attached whole plants and (2) that under conditions of high water velocity it may be easier to feed on drift rather than on attached algae. Hence, the factors of attachment and accessibility will play an important role in determining consumption rates for abalone that predominantly drift-feed. Future research on abalone foraging behavior must further examine these factors in the field quantitatively and with manipulative experiments, to test if these hypotheses are correct, to determine the extent that specific abalone species feed on drift algae, and to determine if these species are capable of algal removal in the field.

ACKNOWLEDGMENTS

The authors thank VUW and VUCEL, for providing laboratory space and access to the resources used to conduct this study; J. Aguirre-Davies, 1. Van de Ven, A. Fleming, and K. Beveridge; for help with the collection of algae and abalone, and J. Aguirre-Davies, C. Hepburn, R. Taylor, and J. Bell; for comments on earlier drafts of this manuscript.

LITERATURE CITED

Alcantara, L. B. & T. Noro. 2005. Effects of macroalgal type and water temperature on macroalgal consumption rates of the abalone Haliotis diversicolor Reeve. J. Shellfish Res. 24:1169-1177.

Byrnes, J., J. J. Stachowicz, K. M. Hultgren, A. R. Hughes, S. V. Olyarnik & C. S. Thornber. 2006. Predator diversity strengthens trophic cascades in kelp forests by monitoring herbivore behaviour. Ecol. Lett. 9:61-71.

Choat, J. H. & D. R. Schiel. 1982. Patterns of distribution and abundance of large brown algae and invertebrate herbivores in subtidal regions of Northern New Zealand. J. Exp. Mar. Biol. Ecol. 60:129-162.

Cronin, G. & M. E. Hay. 1996. Susceptibility to herbivores depends on recent history of both the plant and animal. Ecology 77:1531-1543.

Day, E. G. & G. M. Branch. 2002. Influences of the sea urchin Parechinus angulosus (Leske) on the feeding behaviour and activity rhythms of juveniles of the South African abalone Haliotis midae Linne. J. Exp. Mar. Biol. Ecol. 276:1-17.

Duggins, D. O. & J. E. Eckman. 1997. Is kelp detritus a good food for suspension feeders? Effects of kelp species, age and secondary metabolites. Mar. Biol. 128:489-495.

Dutton, S. & Tong, L. J. 1981. Food preferences of paua. Catch 15-16. Estes, J. A., D. R. Lindberg & C. Wray. 2005. Evolution of large body size in abalones (Haliotis): patterns and implications. Paleobiology 31:591-606.

Estes, J. A. & P. D. Steinberg. 1988. Predation, herbivory, and kelp evolution. Paleobiology 14:19-36.

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

Fletcher, W. J. 1987. Interactions amongst subtidal Australian sea urchins, gastropods, and algae: effects of experimental removals. Ecol. Monogr. 57:89-109.

Gagnon, P., L. Vincent St-Hilaire-Gravel, J. H. Himmelmann & L. E. Johnson. 2006. Organismal defences versus environmentally mediated protection from herbivores: unravelling the puzzling case of Desmarestia viridis (Phaeophyta). J. Exp. Mar. Biol. Ecol. 334:10-19.

Geiger, D. L. 2000. Distribution and biogeography of the Haliotidae (Gastropoda: Vetigastropoda) world-wide. Boll. Malacol. 35:57-120.

Guest, M. A., P. D. Nichols, S. D. Frusher & A. J. Hirst. 2008. Evidence of abalone (Haliotis rubra) diet from combined fatty acid and stable isotope analyses. Mar. Biol. 153:579-588.

Guzman del Proo, S. A., E. Serviere-Zaragosa & D. Siqueiros-Beltrones. 2003. Natural diet of juvenile Haliotis fulgens and H. corrugata (Mollusca: gastropoda) in Bahia Tortugas, Mexico. Pac. Sci. 57:319-324.

Hurd, C. L. 2000. Water motion, marine macroalgal physiology, and production. J. Phycol. 36:453-472.

Kawamata, S. 1998. Effect of wave-induced oscillatory flow on grazing by a subtidal sea urchin Strongylocentrotus nudus (A-Agassiz). J. Exp. Mar. Biol. Ecol. 224:31-48.

Keesing, J. & F. E. Wells. 1989. Growth of the abalone Haliotis roei Gray. Aust. J. Mar. Freshwater Res. 40:199-204.

Keselman, H. J., J. Algina & R. K. Kowalchuk. 2001. The analysis of repeated measures designs: A review. Br. J. Math. Stat. Psychol. 54:1-20.

Littler, M. M. & D. S. Littler. 1980. The evolution of thallus form and survival strategies in benthic marine macroalgae: field and laboratory tests of a functional form model. Am. Nat. 116:25-44.

Lubchenco, J. 1978. Plant species diversity in a marine intertidal community: importance of herbivore food preference and algal competitive abilities. Am. Nat. 112:23-39.

Lubchenco, J. & S. D. Gaines. 1981. A unified approach to marine plant-herbivore interactions. I. Populations and communities. Annu. Rey. Ecol. Syst. 12:405-437.

Marsden, I. D. & D. M. J. Williams. 1996. Factors affecting the grazing rate of the New Zealand abalone Haliotis iris Martyn. J. Shellfish Res. 15:401-406.

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

McShane, P. E., M. G. Smith & K. H. H. Beinssen. 1988. Growth and morphometry in abalone (Haliotis rubra) from Victoria. Aust. J. Mar. Freshwater Res. 39:161-166.

Miner, C. M., J. M. Alstatt, P. T. Raimondi & T. E. Minchinton. 2006. Recruitment failure and shifts in community structure following mass mortality limit recovery prospects of black abalone. Mar. Ecol. Prog. Ser. 327:107-117.

Norderhaug, K. M., S. Fredriksen & K. Nygaard. 2003. Trophic importance of Laminaria hyperborea to kelp forest consumers and the importance of bacterial degradation to food quality. Mar. Ecol. Prog. Ser. 255:135-144.

Paul N. A., R. de Nys & P. D. Steinberg. 2006. Seaweed-herbivore interactions at a small scale: direct tests of feeding deterrence by filamentous algae. Mar. Ecol. Prog. Ser. 323:1-9.

Pearse, J. S. & V. B. Pearse. 1973. Removal of glycine from solution by the sea urchin Strongylocentrous purpuratus. Mar. Biol. 19:281-284.

Peterson, C. H. & P. E. Renaud. 1989. Analysis of feeding preference experiments. Oecologia 80:82-86.

Poore, G. C. B. 1972. Ecology of New Zealand Abalones, Haliotis species (Mollusca: Gastropoda) 1. Feeding. N. Z. J. Mar. Freshw. Res. 6:11-22.

Prince, J. S., W. G. LeBlanc & S. Macia. 2004. Design and analysis of multiple choice feeding preference data. Oecologia 138:1-4.

Rothausler, E. & M. Thiel. 2006. Effect of detachment on the palatability of two kelp species. J. Appl. Phycol. 18:423-435.

Sainsbury, K. J. 1982. Population dynamics and fishery management of the paua, Haliotis iris. I. Population structure, growth, reproduction, and mortality. N. Z. J. Mar. Freshw. Res. 16:147-161.

Schiel, D. R. 1982. Selective feeding by the echinoid Evechinus chloroticus, and the removal of plants from the subtidal algal stands in northern New Zealand. Oecologia 54:379-388.

Shears, N. T. 2007. Biogeography, community structure and biological habitat types of subtidal reefs on the South Island West Coast, New Zealand. Science for Conservation 281:53.

Shears, N. T. & R. C. Babcock. 2002. Marine reserves demonstrate top-down control of community structure on temperate reefs. Oecologia 132:131-142.

Shears, N. T. & R. C. Babcock. 2003. Continuing trophic effects after 25 years of no-take marine reserve protection. Mar. Ecol. Prog. Ser. 246:1-16.

Shears, N. T. & R. C. Babcock. 2007. Quantitative description of mainland New Zealand's shallow subtidal reef communities. Science for Conservation 280:126.

Shepherd, S. A. 1973. Studies on Southern Australian abalone (Genus Haliotis).1. Ecology of five sympatric species. Aust. J. Mar. Freshwater Res. 24:217-257.

Shepherd, S. A. & P. D. Steinberg. 1992. Food preferences of three Australian abalone species with a review of the algal food of abalone. In: Shepherd, S. A., M. J. Tegner & S. Guzman del Proo, editors. Abalone of the world: biology, fisheries, and culture. Oxford: Blackwell. pp. 169-181.

Sinclair, M. 1963. Studies on the paua, Haliotis iris Martyn, in the Wellington district. 1945-1946. Zoology publications from Victoria University of Wellington 35:1-16.

Steinberg, P. D., J. A. Estes & F. C. Winter. 1995. Evolutionary consequences of food chain length in kelp communities. Proc. Natl. Acad. Sci. USA 92:8145-8148.

Steinberg, P. D. & I. Van Alterna. 1992. Tolerance of marine invertebrate herbivores to brown algal phlorotannins in temperate Australasia. Ecol. Monogr. 62:189-222.

Steneck, R. S. & L. Watling. 1982. Feeding capabilities and limitations of marine herbivorous molluscs: A functional group approach. Mar. Biol. 68:299-319.

Stuart-Smith, R. D., N. S. Barrett, C. M. Crawford, S. D. Frusher, D. G. Stevenson & G. J. Edgar. 2008. Spatial patterns in impacts of fishing on temperate rocky reefs: Are fish abundance and mean size related to proximity to fisher access points? J. Exp. Mar. Biol. Ecol. 365:116-125.

Tahil, A. S. & M. A. Juinio Menez. 1999. Natural diet, feeding periodicity and functional response to food density of the abalone Haliotis asinina L. (Gastropoda). Aquacult. Res. 30:95-107.

Taylor, R. B. & P. D. Steinberg. 2005. Host use by Australasian seaweed mesograzers in relation to feeding preferences of larger grazers. Ecology 86:2955-2967.

Tegner, M. J. & P. K. Dayton. 2000. Ecosystem effects of fishing in kelp forest communities. ICES J. Mar. Sci. 57:579-589.

Tegner, M. J. & L. A. Levin. 1982. Do sea urchins and abalone compete in California kelp forest communities? International Echinoderms Conference. Tampa Bay.

Tunbridge, B. R. 1967. Feeding habits of paua. Fisheries Technical Report. N.Z. marine department 20:1-18.

Tutschulte, T. C. & J. H. Connell. 1988. Feeding behaviour and algal food of three species of abalone (Haliotis) in Southern California. Mar. Ecol. Prog. Ser. 49:57-64.

Underwood, A. J. 1981. Techniques of analysis of variance in experimental marine biology and ecology. Oceanogr. Mar. Biol. Annu. Rey. 19:513-605.

Wood, A. D. & C. D. Buxton. 1996. Aspects of the biology of the abalone Haliotis midae (Linne, 1758) on the east coast of South Africa. 1. Feeding biology. S. Afr. J. Mar. Sci. 17:61-68.

CHRISTOPHER E. CORNWALL, * ([dagger]) NICOLE E. PHILLIPS AND DOUG C. MCNAUGHT [(double dagger])

Victoria University, Coastal Ecology Laboratory, School of Biological Sciences, P.O. Box 6002, Victoria University of Wellington, Wellington, New Zealand

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

([dagger]) Current address: Botany Department, P.O. Box 56, University of Otago, Dunedin, New Zealand.

([double dagger]) Current address: University of Maine at Machias, Machias, Maine 04654.

TABLE 1.
Significant pairwise comparisons of amount of algae eaten per abalone
per day. Each row is a species pair, (n = 6 trials for each pair),
with the mean amount of algae (wet weight) consumed of each species
(per abalone, per day) [+ or -] s.e., and the P value of the t-test.
Species 1 is always the member of the pair that was consumed in a
greater amount.

                                                     Mean Algae
Division of Species 1        Species 1         Consumed (g) Species 1

  Phaeophyta
                        L. vareigata           0.5917 [+ or -] 0.2586
                        L. vareigata           0.5383 [+ or -] 0.2922
                        L. vareigata           0.5817 [+ or -] 0.2650
                        L. vareigata           0.7467 [+ or -] 0.1791
                        L. vareigata           0.3200 [+ or -] 0.2171
                        Cy. retroflexa         0.2950 [+ or -] 0.2266
                        Cy. retroflexa         0.3133 [+ or -] 0.2422
                        Cy. retroflexa         0.1633 [+ or -] 0.1763
                        Cy. retroflexa         0.6567 [+ or -] 0.1984
                        Z. turneriana          0.3833 [+ or -] 0.2300
                        Z. turneriana          0.4217 [+ or -] 0.2008
                        Z. turneriana          0.7233 [+ or -] 0.1534
                        Car. maschalocarpum    0.7833 [+ or -] 0.1748
                        Car. maschalocarpum    0.4200 [+ or -] 0.2493
                        E. radiata             0.5600 [+ or -] 0.2783
  Chlorophyta
                        Ulva sp.               0.7700 [+ or -] 0.2300
                        Ulva sp.               0.4767 [+ or -] 0.2316
                        Ulva sp.               0.3300 [+ or -] 0.2146
                        C. flexilis            0.3250 [+ or -] 0.2101
                        C. flexilis            0.4300 [+ or -] 0.1683
  Rhodophyta
                        G. circumcincta        0.4450 [+ or -] 0.2490
                        A. armata              0.5783 [+ or -] 0.1959
                        A. armata              0.4560 [+ or -] 0.2049
                        A. armata              0.2040 [+ or -] 0.2158
                        R. gunnii              0.3917 [+ or -] 0.1951
                        Co. officinalis        0.1820 [+ or -] 0.1116
                        Co. officinalis        0.3060 [+ or -] 0.2464

                                                     Mean Algae
Division of Species 1        Species 2         Consumed (g) Species 2

  Phaeophyta
                        C. brownii             0.1167 [+ or -] 0.2897
                        C. flexilis           -0.1933 [+ or -] 0.0961
                        P. lucida              0.0300 [+ or -] 0.2441
                        R. gunnii              0.2677 [+ or -] 0.2292
                        G. circumcincta       -0.1000 [+ or -] 0.08618
                        C. brownii            -0.0250 [+ or -] 0.1533
                        P. lucida             -0.1167 [+ or -] 0.1159
                        G. circumcincta       -0.2333 [+ or -] 0.0961
                        Co. officinalis        0.0217 [+ or -] 0.0822
                        C. geminata           -0.0200 [+ or -] 0.0616
                        P. lucida              0.1650 [+ or -] 0.1702
                        G. circumcincta        0.2450 [+ or -] 0.2127
                        C. flexilis            0.0500 [+ or -] 0.2157
                        P. lucida             -0.1000 [+ or -] 0.1546
                        R. gunnii             -0.1250 [+ or -] 0.0907
  Chlorophyta
                        C. brownii             0.0067 [+ or -] 0.1155
                        C. flexilis            0.2100 [+ or -] 0.1593
                        G. circumcincta       -0.1067 [+ or -] 0.0542
                        G. circumcincta        0.2133 [+ or -] 0.2598
                        Co. officinalis        0.1150 [+ or -] 0.2171
  Rhodophyta
                        C. brownii             0.1050 [+ or -] 0.1535
                        C. brownii            -0.0017 [+ or -] 0.1582
                        P. lucida             -0.0580 [+ or -] 0.2397
                        G. circumcincta        0.0820 [+ or -] 0.2350
                        C. flexilis            0.0200 [+ or -] 0.1519
                        Ulva sp.              -0.1800 [+ or -] 0.1338
                        R. gunnii              0.1020 [+ or -] 0.1591

Division of Species 1         P value

  Phaeophyta
                              0.0280
                              0.0050
                              0.0215
                              0.0147
                              0.0098
                              0.0480
                              0.0033
                              0.0342
                              0.0019
                              0.0087
                              0.0499
                              0.0150
                              0.0014
                              0.0107
                              0.0147
  Chlorophyta
                              0.0026
                              0.0466
                              0.0114
                              0.0398
                              0.0015
  Rhodophyta
                              0.0322
                              0.0003
                              0.0150
                              0.0367
                              0.0434
                              0.0019
                              0.0309