Directory UMM :Data Elmu:jurnal:J-a:Journal of Experimental Marine Biology and Ecology:Vol254.Issue1.Nov2000:

(1)

254 (2000) 85–107

www.elsevier.nl / locate / jembe

Feeding, growth, and fecundity of Abarenicola pacifica in

relation to sediment organic concentration

Debra L. Linton , Gary L. Taghon

Institute of Marine and Coastal Sciences, Rutgers, The State University of New Jersey, 71 Dudley Road,

New Brunswick, NJ 08901-8521, USA

Received 22 October 1999; received in revised form 17 July 2000; accepted 24 July 2000

Abstract

For marine deposit-feeding invertebrates, the distribution of species with different life history strategies has long been known to be correlated with sediment organic concentration. Large populations of opportunistic species are found in sediments with enriched organic concentration, while equilibrium species populate low organic concentration sediments. Differences in both behavioral (e.g. feeding rate) and physiological (e.g. growth rate, reproductive output) adaptations determine the ability of species to establish populations in different environments. By sys-tematically documenting differences in the way these factors vary as sediment organic con-centration varies for both opportunistic and equilibrium species, we can better understand the mechanisms underlying this correlation between sediment organic concentration and species distributions. Here, we present the results of experiments examining the interactions among food concentration, feeding rate, growth rate, and reproductive output (measured as egg number and size) for the equilibrium species Abarenicola pacifica. A. pacifica is a large, long-lived, iteroparous, sub-surface deposit-feeding polychaete. Individual worms were reared throughout most of one generation in sediments differing only in the concentration of organic matter. Juveniles (,20 mg AFDW) had higher feeding rates and growth rates in sediments of higher organic concentration throughout the range tested. These results are consistent with the predictions from optimal foraging theory. As worms grew, however, these patterns changed. Once worms reached a mean body size of |50 mg AFDW, feeding rate was greater on sediments of lower

organic concentration (although it took worms in the sediments with lower organic concentration longer to reach this size). Differences in growth rates among treatments decreased as worms grew. For worms .100 mg AFDW, growth rates were uniformly low (¯_{1% / day) on all sediments, but}

the early advantage obtained by worms in the high organic treatments resulted in much greater body sizes after 200 days. Worms had higher tissue triacylglyceride concentrations and produced more eggs (independent of worm size) as sediment organic concentration increased. We conclude that A. pacifica alters its feeding rate in response to variations in food resources in such a way as

*Corresponding author. Tel.:11-732-932-6555; fax:11-732-932-8578.

E-mail address: [email protected] (D.L. Linton).

(2)

to maximize its energy intake and thereby maximize fitness. Future studies should investigate whether opportunistic species (as well as other equilibrium species) also have this ability.

Keywords: Abarenicola pacifica; Equilibrium species; Functional response; Optimal foraging theory; Poly-chaetes; Sediment organic concentration

1. Introduction

Perhaps the most basic goal of ecology is to explain the observed distribution and abundance of organisms. The suite of adaptations that determines where a species can successfully establish has been called a life history strategy (Stearns, 1976; Grime, 1977). In the study of the marine benthos, classification of life history strategies has been driven by studies of community succession following disturbance (e.g. Grassle and Grassle, 1974; McCall, 1977; Pearson and Rosenberg, 1978). Species that rapidly colonize disturbed areas have strategies that enable them to achieve explosive local population growth and maintain dominance in organically enriched areas. These characteristics include rapid individual growth, early maturity, brood protection, production of lecithotrophic larvae, and production of many broods per year. These species have been termed ‘opportunistic’ species and include small, surface or near-surface deposit-feeding polychaetes. As a site recovers, populations of these opportunists rapidly decline and a community of larger, subsurface deposit-feeding and suspension-feeding species becomes established. These species characteristically are slower growing, have few broods per year (or reproduce annually), produce planktotrophic larvae, and have no brood protection. Because this stage of succession has been called community equilibrium, these species have been termed ‘equilibrium’ species. The key feature of this pattern is that as the environment changes, the adaptiveness, or fitness, of the different life history strategies changes, leading to a change in species composition at the site. Therefore, by examining how fitness characteristics (e.g. fecundity and egg size) of the resident species change over an environmental gradient, we might gain insight into some of the mechanisms affecting species distributions.

One environmental factor that clearly plays a major role in determining the distribution of marine benthic species is the concentration of organic matter in the sediments. A review of relevant experiments by Thistle (1981) led him to conclude that the disappearance of opportunists during succession likely results from the exhaustion of a resource rather than interspecific interactions with the later-arriving equilibrium species. An extensive set of laboratory experiments and field data has shown that individual growth, reproductive rates, and population dynamics of opportunistic deposit-feeding polychaetes are significantly influenced by levels of nutrition (Tenore, 1977,

1983; Chesney and Tenore, 1985; Tenore and Chesney, 1985; Gremare et al., 1988, 1989a,b; Forbes and Lopez, 1990; Tsutsumi et al., 1990; Bridges et al., 1994; Bridges, 1996). In fact, reproduction of some opportunistic species may be restricted to areas where sediments contain some threshold concentration of organic matter (Tsutsumi, 1990; Tsutsumi et al., 1990).

(3)

Due to the experimental difficulties inherent in working with larger, longer-lived species (e.g. the need for large volumes of treatment sediment and difficulties controlling experimental conditions over long periods of time), few laboratory experiments have been done on population dynamics of equilibrium species. Some equilibrium species have been shown to grow more quickly in sediments with higher concentrations of organic matter (e.g. Taghon and Greene, 1990), but no experiments have been conducted to document the effects of differences in concentration of organic matter on reproductive output and population parameters.

By examining both behavioral (food / energy acquisition) and physiological (growth rate, reproductive output) adaptations in species with different life history strategies, we believe that patterns may emerge which will help explain the observed distribution of species in the marine soft-bottom benthos. Therefore, the goal of this study was to measure feeding rate, growth rate, and reproductive output of an equilibrium species over a range of sediment organic matter concentrations. The species selected for this study was the deposit-feeding polychaete Abarenicola pacifica. These experiments are the first to systematically measure feeding, growth, and reproduction of an equilibrium species reared in the laboratory on treatment sediments throughout most of its generation time (from early juvenile stage to first reproduction).

2. Materials and methods

2.1. Study species

Abarenicola pacifica (Healy and Wells, 1959) is a large, subsurface deposit-feeding polychaete, common in estuaries and coastal sediments in the North Pacific Ocean. Adults reach lengths of 8–10 cm and dry weights of 250–300 mg. Individual growth rates are low, reaching a maximum of only |4–5% / day during juvenile growth. Individuals live head-down in J-shaped burrows, ingesting sediment at the end of the horizontal gallery. Worms irrigate their burrow in a headward direction while feeding, ceasing as they crawl backward to deposit feces on the sediment surface. Knowledge about the life cycle of A. pacifica is incomplete. Sexes are listed as separate (Healy and Wells, 1959) without any reference to specific investigations showing this to be strictly true. Broods are produced annually. Eggs are |160–190 mm in diameter, but characteristic brood sizes are not cited in the literature. Embryos are brooded in the burrow and larvae are ‘non-feeding and only briefly pelagic’ (Strathmann, 1987) or possibly totally benthic (Wilson, 1981). Although the possession of benthic larvae and brood protection are more characteristic of opportunistic species, A. pacifica is classified as an equilibrium species. Lugworms (arenicolid polychaetes) have several distinguish-ing features that lead to their classification as equilibrium species. These include their large size, slow growth rate, long generation time (1–2 years), sub-surface habitat, and relatively stable population sizes (Newell, 1948; Hobson, 1967; Longbottom, 1970;

(4)

2.2. Worm collection

Worms for all experiments were collected from the intertidal zone at Yaquina Bay, Oregon, and shipped to our laboratory in New Jersey by overnight courier. Worms always arrived in good condition and were immediately placed in holding sediment in a seawater table (158C, 30–32‰ seawater, recirculating with continuous aeration) pending the start of the experiment.

2.3. Sediment preparation

Sediments for the experiments were collected from the intertidal mudflat at Belmar, New Jersey and the intertidal marsh surface at Tuckerton, New Jersey, in July and August, 1995. Only sediment from the top 5–10 cm was used. Belmar sediment is a moderately rich, muddy sand. Tuckerton sediment is an organic-rich marsh mud that was used to enrich the Belmar sediment. Both sediments were initially sieved through a 1-mm mesh screen to remove large organisms, shell fragments, and plant detritus. Portions of both sediments (Belmar and Tuckerton) were treated with regular household bleach (4% sodium hypochlorite) to remove organic matter. The bleach was neutralized with sodium thiosulfate (10% solution) and the sediment was then rinsed two times with seawater (30‰). Samples of the four sediment types (Belmar, bleached Belmar, Tuckerton, and bleached Tuckerton) were collected for chemical analyses.

These four sediment types were mixed in different proportions to create sediments of varying organic concentrations (Table 1). The overall proportions of Belmar (bleached1unbleached) and Tuckerton (bleached1unbleached) sediments were held constant among treatments to minimize differences in grain size distributions. Each treatment was composed of 90% Belmar sediment and 10% Tuckerton sediment. Varying the percentage of bleached and unbleached sediment within that 90% and 10% created the different treatments (Table 1). Approximately 75 l of each treatment was prepared by mixing the components in a large cement mixer for 30 min. The sediments were frozen in 7-l batches at 2208C and were thawed and rinsed (two times) with seawater prior to use. Sediment samples for chemical analyses were collected each time a new 7-l

Table 1

Sediment mixtures for Abarenicola pacifica experiments

Treatment Belmar Belmar Tuckerton Tuckerton

(%) bleached (%) bleached

(%) (%)

25% Belmar (25% B) 25 65 – 10

50% Belmar (50% B) 50 40 – 10

75% Belmar (75% B) 75 15 – 10

90% Belmar (90% B) 90 – – 10

2.5% Tuckerton (2.5% T) 90 – 2.5 7.5

5% Tuckerton (5% T) 90 – 5 5

7.5% Tuckerton (7.5% T) 90 – 7.5 2.5

(5)

aliquot of sediment was thawed and rinsed. These samples were frozen, freeze-dried, and homogenized by grinding. Total nitrogen (TN) and total organic carbon (TOC) were measured using a Carlo Erba NA-1500 elemental analyzer (Valencia, CA, USA). Sediments were first exposed to HCl fumes for 24 h to remove any inorganic carbon (e.g. shell fragments). Protein concentration was measured using the Enzyme Hydrolyz-able Amino Acids technique of Mayer et al. (1995). Sediments were incubated with protease (following the EHAA procedure) for 6 h and total hydrolyzed amino acid concentration was measured spectrofluorometrically. This technique gives a biologically relevant measure of sediment protein concentration because it mimics the action of digestive enzymes and measures only that portion of organic matter that is available to be digested.

2.4. Worm measurement

To estimate body size, worms were anaesthetized in a 3% MgCl solution and their2

planar area was measured by video image analysis. Spherical volume was then calculated and a regression equation was used to estimate ash-free dry weight (AFDW). To develop this equation, Abarenicola pacifica were held in seawater until feces were voided, then their body volumes were estimated, as above. The worms were subsequent-ly dried, weighed, heated at 5008C for 5 h, and re-weighed. A power regression, correlating volume to AFDW, was fit to the data using the statistical softwareSTATVIEW (SAS Institute, 1998). The equation generated was:

0.932 2

AFDW527.22*volume (R 50.978, n572).

Relative growth rates (RGR, % / day) were calculated (based on the estimated AFDWs) as in Radford (1967).

2.5. Apparatus

All experiments were conducted in a seawater table with a continuous flow of recirculated seawater and continuous aeration to prevent oxygen depletion. Temperature was maintained at 158C for all experiments. This temperature is toward the upper end of the range that Abarenicola pacifica experiences in situ. Salinity was measured approximately every 2 days and maintained between 30 and 32‰ by addition of fresh water as needed to compensate for evaporation. For all experiments, individual worms were placed in plastic containers (7.5 cm tall, 11 cm top diameter, 8 cm bottom diameter) that were filled with treatment sediment (|7 cm deep) and randomly placed in the seawater table to control for any position effects.

2.6. Experiment 1

Juvenile worms were collected in late July 1995. Based on the timing of spawning of Abarenicola pacifica in Yaquina Bay, we estimate that these worms were 2–3 months old. The worms were held in diluted Belmar sediment prior to the start of the experiment

(6)

Table 2

a Sampling period dates for Abarenicola pacifica experiments

Period Dates Duration (days)

Experiment1

Period 1 10 / 6–11 / 11 / 95 36

Period 2 11 / 12–12 / 14 / 95 34

Period 3 12 / 15 / 95–01 / 18 / 96 55

Period 4 01 / 19–02 / 21 / 96 44

Period 5 02 / 22 –03 / 21 / 96 29

Total 198

Experiment2 11 / 04 / 97–05 / 04 / 98 181

Sediments were not used in Experiment 2.

to keep growth to a minimum. On September 6, worms were measured and then placed singly into each of 15 sediment containers per treatment.

At the start of each sampling period (Table 2), five worms per treatment were randomly selected for feeding rate measurement. All feces were collected from these worms every 2–3 days throughout the sampling period. At the same time, feces were also cleared from all other worms. The collected feces were rinsed with distilled water to remove salts and then dried and weighed. An estimate of daily feeding rate for each worm was calculated by dividing the total mass of feces collected by the number of days in each sampling period. Specific feeding rates were calculated by dividing this average mass of feces produced per day by the geometric mean size of the worm (mg estimated AFDW of the worm at the midpoint of the sampling period).

Worms were sieved from the sediment approximately monthly (Table 2) and survival and size were recorded. The worms were also assessed for evidence of gametogenesis by observation under a dissecting microscope with darkfield illumination. The worms were then replaced in freshly thawed and rinsed treatment sediment in the seawater table. The experiment was continued until the presence of eggs in all of the surviving worms indicated that spawning might be imminent. The duration of the experiment was 198 days.

At the conclusion of the experiment, the worms were measured and then held in seawater to void feces. The worms were frozen at2808C and subsequently freeze-dried and weighed. Worm tissue was homogenized by grinding and analyzed for the presence of glycogen, protein, and triacylglycerides. Lipids and glycogen are the major energy storage products in aquatic invertebrates (Giese, 1966). Taghon et al. (1994) have shown that large increases in triacylglyceride levels are associated with vitellogenesis by Abarenicola pacifica and suggest the measurement of triacylglyceride levels as an objective method of quantifying reproductive effort in deposit feeders.

Tissue glycogen was measured by the method of Keppler and Decker (1984) and tissue protein was measured using Pierce’s BCA Protein Assay Kit (Product[ _23225).

Tissue triacylglycerides were extracted in a solvent of 3:2 hexane:isopropanol, following the procedure of Radin (1981). This solvent was chosen because it extracts almost no protein and non-lipid material, including pigments, which are extracted by traditional solvents. Also this solvent has a lower density which permits centrifugation as an

(7)

alternative to filtration. After triacylglycerides were extracted, the solution was cen-trifuged to pellet the insoluble portions and the liquid portion was transferred to another test tube. The extraction was then repeated. The solvent was evaporated at 508C under N2 to dryness and 0.1 ml of solvent was added to standardize volume. Selective hydrolysis of the triacylglycerides into glycerol was performed in E NOH (tetra-ethyl4

ammonium hydroxide) following the method of Chernick (1969). Glycerol concen-trations were measured spectrophotometrically using Sigma’s Triglyceride Diagnostic Kit (Product [ _{337-B; Procedure} [_{337) based on the method of McGowan et al.}

(1983).

2.7. Experiment 2

Juvenile worms were collected in late October 1997 and were estimated to be 5–6 months old. In November, worms were placed singly in the treatment sediment containers. Four worms were tested in each of six sediment types (75% B–10% T).

Fecal castings were removed and sediment was added as needed to the containers, but the worms were not disturbed until March 1998. From this time on, the worms were sieved out and checked for eggs monthly. On day 181 (May 4), eggs were present in all of the worms, indicating that spawning might be imminent. Each worm was then preserved in buffered formalin, measured, dissected, and the eggs were collected. The

eggs were evenly dispersed in a gridded petri dish (area574 cm ). Eggs were counted in

ten randomly selected grids (each grid50.3 cm30.275 cm, area50.0825 cm ) and an estimate of the total number of eggs per worm was calculated. The diameters of 100 randomly selected eggs were measured (by ocular micrometer at 703 magnification) for each worm and these measurements were averaged to give an estimate of egg size for each worm. Lugworm eggs develop a distinct bi-concave shape during advanced development (Howie, 1961). Egg measurements were marked to distinguish those in this more advanced developmental stage from those in early development.

2.8. Statistical analysis

All statistics were performed using STATVIEW version 5.0 (SAS Institute, 1998) and SUPERANOVA version 1.11 (Abacus Concepts, 1991). Individual worms were treated as replicates for all analyses. Treatment effects on worm size, relative growth rate, specific feeding rate, tissue chemistry, and egg size were compared using one- or two-way analysis of variance (ANOVA). Feeding rates and egg counts were analyzed using analysis of covariance (ANCOVA) in order to separate the effects of worm size from the treatment effects. An alpha level of 0.05 was used as the criterion for rejecting the null hypothesis in all tests. Geometric mean size of each worm during a sampling period was used as the covariate in all feeding rate analyses. Estimated final AFDW of each worm was used as the covariate for analysis of egg counts. The ANCOVA assumption of homogeneity of regression slopes (dependent variable vs. covariate) was tested following the method of Huitema (1980). All ANCOVA analyses presented fit the assumption of homogeneity of regression slopes. The assumption of homogeneity of variances was examined for all analyses by plotting residuals versus fitted values. If heterogeneity was

(8)

found, data were transformed to stabilize the variances. Transformations are indicated in the figure legends.

When significant main effects were detected in ANOVA analyses, multiple com-parisons of treatment means were performed using Fisher’s protected least significant difference (LSD) test. Multiple comparisons of ANCOVA-adjusted treatment means were performed using the least squares method.

Smoothed regression lines were fit to bivariate scatterplots using Cleveland (1979, 1981) LOcally-WEighted Scatterplot Smoother (LOWESS). This is a robust procedure that diminishes the distorting effect of outliers by assigning greater weight to locally grouped data points. LOWESS fits were performed inSTATVIEW, using the default tension setting of 66%.

3. Results

3.1. Sediment chemistry

There was considerable variability among replicate measures of sediment total organic carbon, total nitrogen, and protein (Table 3). Most of the variability can be ascribed to two causes. The first is the freezing, thawing, and rinsing process performed for each batch of sediment, which may have resulted in the differential loss of fine, organically rich particles. A second possibility is that the sediment samples taken were not true representatives of the bulk sediment. Sediments are notoriously heterogeneous and it is difficult to obtain a representative measure by analyzing 0.1 g replicates taken from an |10-ml sample from a 7-l batch of treatment sediment. Replicate analyses of sediments from the same sample showed high variation, supporting this explanation. Nonetheless, there was a clear increase in total organic carbon for all treatments. Protein concentration increased through the first four treatments, in which the amount of bleached Belmar sediment was varied, then showed an unexpected, non-monotonic, pattern in the next four treatments. However, all Tuckerton-supplemented mixtures had significantly higher protein concentrations than the nearest Belmar treatment.

Table 3

a Chemical properties of sediments used in Abarenicola pacifica Experiments 1 and 2

Treatment TOC (mg / g) (n58) TN (mg / g) (n58) Protein (mg / g) b

25% Belmar 1.9761.55 0.25260.113 0.21260.168 (n526) b

50% Belmar 2.5061.49 0.22460.108 0.24360.186 (n538) 75% Belmar 3.5961.60 0.30860.115 0.39460.194 (n548) 90% Belmar 4.1561.44 0.33860.099 0.41760.196 (n547) 2.5% Tuckerton 5.6160.410 0.44460.045 0.58660.132 (n548) 5% Tuckerton 5.7760.301 0.48660.062 0.62260.170 (n550) 7.5% Tuckerton 6.1860.979 0.46260.092 0.53260.190 (n550) 10% Tuckerton 6.2460.360 0.50260.079 0.58360.204 (n550)

Data are means6S.D., n5number of replicates. b

(9)

3.2. Feeding rate

During the first sampling period, specific feeding rates (g sediment / g AFDW/ day) were higher for worms feeding on sediments with higher organic concentrations (Fig. 1A). ANCOVA-adjusted means (which remove the effect of differences in body size) of bulk feeding rate (g / day) also increased with increasing treatment mix percentages (75% B,90% B,2.5% T,5% T, 7.5% T,10% T) (Fig. 1B). The 25% B treatment was excluded from these analyses because only two of the worms that were randomly selected for feeding measurements survived to the end of the period. LOWESS regressions of specific feeding rates vs. geometric mean size for the entire duration of the experiment show a more complex pattern (Fig. 2A). Specific feeding rates in the Tuckerton treatments were higher in juveniles, but decreased with increasing worm size. In the 90% and 75% Belmar treatments, specific feeding rates were initially lower, but increased as worm size increased up to |50 mg AFDW, after which feeding rates followed the same track as worms in the Tuckerton treatments. Bulk feeding rates increased linearly with worm size throughout the experiment for worms in the Tuckerton sediments, but approached a plateau in the 90% and 75% Belmar treatments (Fig. 2B). For GMS of |40–100 mg AFDW, worms were feeding faster in the 90% and 75% Belmar treatments than were worms in the Tuckerton treatments when they were at that size.

3.3. Growth

Relative growth rates during period 1 were generally higher on sediments with higher organic concentrations. Many worms in the 25% B and 50% B treatments had negative growth rates. Survival was low in these treatments (Fig. 3). By the end of the experiment all of the 25% B worms had died and only one worm remained alive in the 50% treatment. Growth rates were positive in the remaining treatments and showed significant increases between 75% B, 90% B and the Tuckerton-addition sediments (Fig. 4A). Survival was high in these treatments (Fig. 3).

A LOWESS regression plot of RGR vs. GMS over the duration of the entire experiment (Fig. 4B) shows the expected allometric decrease in relative growth rates with increasing age / size and also clearly shows higher growth rates in sediments with higher organic concentration for worms,100 mg AFDW. Above this size, growth rates were low and uniform at ¯_{1.0% / day for all treatments. However, at all sampling dates}

worms in sediments with higher organic concentration were significantly larger due to the head start conferred by higher early growth rates (Fig. 5).

3.4. Tissue chemistry

There were no significant differences among treatments in tissue glycogen (Fig. 6A) or tissue protein (Fig. 6B), however, tissue triacylglyceride levels increased significantly with increasing sediment protein organic concentration (Fig. 6C). Triacylglyceride concentrations were two–three times higher for worms in the Tuckerton treatments.

(10)

Fig. 1. Feeding rate data for Period 1 of Experiment 1. Data are means6S.D. Means grouped by letter are not significantly different. (A) Specific feeding rate (data were ln transformed for ANOVA) and (B) adjusted bulk feeding rate (based on ANCOVA with worm geometric mean size as the covariate).

(11)

Fig. 2. LOWESS regression plots (scatterplot points removed for clarity) of feeding rate vs. GMS for Experiment 1. (A) Specific feeding rate vs. GMS and (B) bulk feeding rate vs. GMS.

(12)

Fig. 3. Survival (%) vs. days for Experiment 1. 3.5. Egg production

In both Experiment 1 and Experiment 2, all worms survived and produced eggs. The low level of replication and high level of variability between worms in Experiment 2 led to low statistical power, but a distinct trend was evident. Egg number (independent of body size) increased significantly with increases in sediment organic concentration (Fig. 7A). Worms in the 75% B treatment produced approximately half as many eggs as worms in the 90% B treatment. There was no significant difference in egg number between the 90% B, 2.5% T, 5% T, and 7.5% T treatments, but worms in the 10% Tuckerton treatment produced 20–30% more eggs than worms in those treatments.

Mean egg diameter increased from 75% B to 90% B to Tuckerton treatments, but these differences were not significant (Fig. 7B). A significantly lower percentage of eggs in the 75% B treatment showed the bi-concave shape associated with advanced development (Fig. 7C). A comparison of the sizes of only eggs which had reached advanced development also showed no significant differences in egg size among treatments.

4. Discussion

4.1. Feeding rate

The observation that deposit feeders can alter their feeding rates in response to changes in sediment organic matter concentration dates back more than 30 years

(13)

Fig. 4. Relative growth rate (RGR) data for Experiment 1. (A) RGR for period 1 by treatment. Data are means6S.D. Means grouped by letter are not significantly different. (B) LOWESS regression of RGR vs. GMS (scatterplot points removed for clarity) for the entire experiment.

(14)

Fig. 5. AFDW over time for Experiment 1. Data are means (S.D. excluded for clarity). Treatments grouped by letter are not significantly different (based on two-way ANOVA of AFDW as a function of sediment treatment and sampling period). AFDW data were ln transformed.

(Gordon, 1966). Subsequent studies have not yielded consistent results. Some species ´

increase their feeding rate as sediment organic concentration increases (Cadee, 1976; de Wilde and Berghuis, 1979; Whitlatch and Weinberg, 1982; Taghon and Jumars, 1984), while others decrease their feeding rate (Nichols, 1974; Kudenov, 1982; Yingst, 1982). Optimal foraging theory has been used to model the shape of the functional response curve (Taghon, 1981; Calow, 1982; Phillips, 1984; Cammen, 1989). The most advanced model for deposit feeders incorporates digestion and absorption kinetics (Dade et al., 1990). A generalized performance analysis of guts, modeled as continuous plug-flow chemical reactors, leads to the prediction that feeding rate should increase as food concentration increases, reaching a maximum at some intermediate concentration, and then decrease as food concentration exceeds this critical value. This pattern of response should maximize the absorption of digestive products (Dade et al., 1990). Since deposit feeders subsist on a nutrient-poor food source and food gathering is such a critical and time-consuming activity for them (Lopez and Levinton, 1987), it is not unreasonable to use this ‘optimal’ model as a hypothesis for testing the relation between feeding rate and sediment organic concentration. The peaked functional response predicted by Dade et al. (1990) has been seen in recent experiments with several taxa. The deposit-feeding amphipod, Corophium volutator, increases feeding rate, followed by a decrease, as concentrations of microalgae in the sediment increase (Cammen, 1989). The egestion rate of a surface deposit-feeding acorn worm, Saccoglossus kowalevskii, peaks at an intermediate sediment chlorophyll a concentration, then gradually declines as con-centration continues to increase (Karrh and Miller, 1994). Abarenicola pacifica also

(15)

Fig. 6. Tissue chemistry data for Experiment 1. (A) Tissue glycogen, (B) tissue protein, and (C) tissue triacylglycerides. Data are means6S.D. Means grouped by letter are not significantly different.

(16)

Fig. 7. Egg data for Experiment 2. Data are means6S.D.; n54 worms per treatment. Means grouped by letter are not significantly different. (A) Adjusted number of eggs produced (based on ANCOVA with worm AFDW as the covariate), (B) egg diameters, and (C) average percentage of eggs showing advanced development (bi-concave shape).

(17)

showed the peaked response of feeding rate over a gradient of sediment protein concentration created by diluting native sediment with clean sand and supplementing native sediment with baby food (Taghon and Greene, 1990; Taghon et al., 1990).

In the present experiment, feeding rate of A. pacifica during the first sampling period was higher in sediments with higher organic matter concentration throughout the range tested. Analyses based on both specific feeding rates (Fig. 1A) and ANCOVA-adjusted feeding rates (Fig. 1B) showed this pattern. Worms were approximately 4–5 months old during this period and signs of reproductive activity were not yet apparent [i.e. no ‘milky’ appearance of coelomic fluid (Newell, 1948)]. These data permit the best direct comparison with results from other studies of the effect of sediment protein con-centration on feeding rate because the worms were initially the same size and were in the same reproductive condition. In Taghon and Greene (1990) and Taghon et al. (1990) experiments, feeding rate (g / day) peaked at a sediment protein concentration of 0.05–0.1 mg / g, decreasing with both increases and decreases in sediment protein concentration. The lowest protein concentration used in the present experiments was |0.2 mg / g. Based on previous results we would have expected to see only decreases in feeding rates as sediment protein concentration increased above this level, but this was not the case. Feeding rates increased as sediment protein concentration increased.

One methodological difference between these experiments is the measurement of sediment protein concentration. Taghon and Greene (1990) and Taghon et al. (1990) used the Coomasie Blue (CB) method (Mayer et al., 1986). The CB method is sensitive only to larger polypeptides (.15–20 amino acids), whereas the EHAA method measures both small and large peptides that are available to hydrolysis by protease. The EHAA method always showed higher yields than the CB method in Mayer et al. (1995) initial testing of a variety of sediments, but this difference is not large enough to account for the conflicting observations.

Another difference between the studies is that the sediments used were very different. The sediments in previous experiments (Taghon and Greene, 1990; Taghon et al., 1990) were much sandier than the Belmar sediments used in these experiments (personal observation). Also, the addition of 10% Tuckerton sediment for each treatment led to sediments with much smaller average grain sizes than those used previously. Differences in grain size distribution may cause differences in the digestion and assimilation of sediment protein. If muddier sediments limit diffusion of digestive enzymes in the gut, the worms may not be able to digest all of the otherwise bioavailable protein and may respond to the sediment as if it had a lower protein concentration. Also, different sources of organic matter may be more or less easily digested, leading to peaks in feeding rates at different absolute protein concentrations.

By continuing to follow feeding rates throughout the life cycle of Abarenicola pacifica, we saw interesting complexities develop as worm sizes and gametogenesis began to differ among treatments. LOWESS regression of specific feeding rates vs. body size for the entire duration of the experiment (Fig. 2A) showed specific feeding rate decreasing with increasing worm size in the Tuckerton sediments, then leveling off for worms greater than 75 mg AFDW. In the 75% and 90% Belmar sediments, however, feeding rates initially increased with increasing worm size up to |50 mg AFDW, then followed the same track as worms in the Tuckerton sediments. Bulk feeding rates

(18)

gradually (and approximately linearly) increased with worm size for the Tuckerton treatments, but increased rapidly, then plateaued, for the 75% and 90% Belmar treatments (Fig. 2B). Worms in the lower quality sediments (75% and 90% Belmar) had higher bulk feeding rates than those in the higher quality sediments (all Tuckerton treatments) once body sizes reached |50 mg AFDW. It may have taken the worms in these treatments longer to reach these body sizes, but they were feeding at higher rates than worms in the richer treatments were when they were at that same size. Some of the different response patterns seen among worms feeding in sediments with different protein organic concentrations may be due to the differential timing of allometric shifts in feeding, growth, and progression of reproduction. From the perspective of comparing these data with model predictions and previous experimental results, these different trajectories would lead to different conclusions depending on the size / age and prior nutritional history of the worms used in separate experiments. Small (and presumably young) worms fed faster in sediments with higher organic concentrations. Larger animals (75–100 mg AFDW) showed the opposite pattern. There are no reliable methods for aging soft-bodied marine invertebrates such as deposit-feeding polychaetes. The usual assumptions are that similarly sized individuals are of similar age and have similar nutritional histories. Errors in these assumptions could lead to opposite conclusions when comparing experiments, even experiments using the same species. Ideally, the age and nutritional history of animals used in experiments should be known, but this is not always practical. It is often much easier to use animals that were recently collected from the field in laboratory experiments. Doing so may introduce unexpected levels of variability, however, making interpretation of results more ambiguous. Due to these complexities, we must concentrate our analysis on differences among treatments within this experiment. The pattern of higher feeding rates in sediments with greater organic concentration is not inconsistent with the optimal foraging model of Dade et al. (1990) for sediments below the ‘threshold’ organic concentration.

4.2. Growth

Growth rate increased as sediment organic concentration increased, reaching a plateau at 5–6% / day in the 5–10% Tuckerton sediments. This plateau of growth rate may indicate that Abarenicola pacifica had reached its maximum potential growth rate or it could indicate that these sediments did not have (biologically) significantly different concentrations of organic matter. By following growth rates throughout the experiment, both allometric patterns and treatment effects were discernible. Not only were growth rates higher in sediments with higher organic concentrations, but the characteristic allometric pattern of decreasing growth rates with increasing age / size (Peters, 1983) was seen. Differences in growth rates among treatments disappeared in worms above |100 mg AFDW (Fig. 4B). Growth rates at this size were ¯1% / day. Nevertheless, worms in the higher quality sediments were always significantly larger and worms in all treatments continued to grow, even while completing vitellogenesis.

These growth rate results are qualitatively consistent with the results seen by Taghon and Greene (1990) and Taghon et al. (1990). In those experiments, growth rates also increased as sediment protein concentration increased. Quantitatively, however, there are

(19)

differences between the previous and present results. At low protein concentrations, growth rates in the previous experiments were much higher. Maximum growth rates in the present experiments were higher than those recorded previously, but the protein concentrations were also higher. As was the case for feeding rate, there are also pitfalls in comparing growth rate among experiments. There was considerable variability in growth and feeding rates between experiments in Taghon and Greene (1990). Worms placed in sediments with protein concentrations between 0.05 and 0.2 mg / g in June 1988 fed three–four times faster and grew approximately four times faster than worms placed in different sediments with the same range of sediment protein concentration in July 1988. Some of the variability in growth rates can be ascribed to worm size or age. Growth rates decrease as body size increases. Another source of variability is, again, the differences in the quality of the organic matter in different sediments.

Although growth rates consistently correlate best with measures of sediment protein concentration (e.g. Tenore, 1977; Marsh et al., 1989; Horng, 1998), there are clearly other properties of sediment that are relevant to deposit feeders. Micronutrients such as essential amino acids have been shown to be critical to growth and reproduction in

Capitella sp. I (Marsh et al., 1989; Gremare, 1994). Tsutsumi et al. (1990) found that Capitella sp. I grew at significantly different rates in two sediments with the same protein concentration (|4 mg / g) in two different experiments. In the experiment with higher growth rates the sediment had been enriched to this level by addition of powdered algae, while the other used native sediment with this protein concentration. The sources of protein in the two experiments were different and worms responded to them differently. Hylleberg (1975) found that while ciliates, flagellates, nematodes, motile bacteria, and some motile diatoms were readily digested by Abarenicola pacifica, bacteria and diatoms attached to sediment grains and detrital fragments passed the gut undigested. All of these sources might be measured as bioavailable protein, but not all may actually be available to deposit feeders. Therefore, different sediments, with their complex mixes of these components, may have the same protein concentration but be of different value to organisms and elicit different functional and physiological responses.

4.3. Reproduction

Worms reared in sediments with higher organic concentration produced more eggs. Because egg sizes were not significantly different among treatments, the increased levels of triacylglycerides in egg-bearing worms reared in sediments with higher organic concentrations can be directly related to increased fecundity. Triacylglycerides are the primary energy storage product in Abarenicola pacifica eggs (Taghon et al., 1994). Glycogen appears not to be involved in egg production in A. pacifica, as indicated by both no significant differences among treatments and random variability within treatments. This result agrees with the findings of Taghon et al. (1994), who also showed that glycogen levels did not vary seasonally or with reproductive condition, as we might expect if glycogen were being used as a storage product prior to gameto-genesis. However, in Arenicola marina, glycogen levels were closely correlated with proliferation of gametes and no relation was apparent between total lipids and the

(20)

reproductive cycle (de Vooys, 1975). Tissue protein concentrations also showed no differences between treatments.

An interesting sidebar is the fact that in both experiments all surviving worms produced eggs. This was unexpected, because Abarenicola pacifica is classified as a dioecious species (Healy and Wells, 1959). Although sex ratio in Arenicola marina has been shown to be skewed in favor of females (3.75: 1) (Newell, 1948), even if the same is true for Abarenicola pacifica, it is very unlikely that only females would be randomly collected in a sample of over 120 worms. Alternatively, only females may have survived. Of the 90 worms in the six highest organic concentration treatments, 67 survived. A detailed study is needed to determine if hermaphroditism is present in Abarenicola pacifica.

All of the previous measurements of reproductive output in polychaetes in relation to food quality have had two major differences from the present study. First, the previous studies have all been done with species with short generation times, and secondly, previous studies have all compared reproduction of worms fed either different types of food (i.e. different protein sources) or different rations (i.e. same concentration, different amounts per day). Most work has been done on the opportunistic species Capitella sp. I. Increases in fecundity have been seen repeatedly in response to higher ‘quality’ foods in

´ ´

Capitella sp. I (Gremare et al., 1988, 1989a; Bridges et al., 1994; Gremare, 1994). Increases in fecundity have also been seen in Capitella sp. I in response to increased

ration (Gremare et al., 1989a,b). Another opportunist, Streblospio benedicti, does not show dramatic changes in fecundity in response to food quality (Bridges et al., 1994). However, the lecithotrophic form of S. benedicti does show increased fecundity with increasing ration (Levin and Creed, 1986). A more recent study (Prevedelli and Zunarelli Vandini, 1998) on the dorvilleid polychaete, Ophyrotrocha labronica, showed much higher fecundity when worms were given a food source with higher protein concentration. O. labronica is a small worm with a short generation time, repeated reproduction, and a high reproductive capacity, characteristics that place it in the opportunistic category.

Measurement of egg sizes produced by worms with different food sources has rarely been done. Bridges (1996) found different organic additions to have a marginal (non-significant) effect on embryo size for Capitella sp. I and suggested that the potential exists for differential per offspring investment. However, Ophyrotrocha labronica had no change in egg size related to different diets (Prevedelli and Zunarelli Vandini, 1998). Similarly, in the present study, egg sizes of Abarenicola pacifica were not significantly different across a range of sediment quality.

5. Conclusions

These are the first long-term measurements of feeding rate, growth rate, and reproductive effort of an equilibrium polychaete species relative to controlled sediment quality. Standing alone as a test of optimal foraging predictions, this study has shown that Abarenicola pacifica altered feeding rates, grew more rapidly, and produced more eggs when feeding on sediments of higher organic concentration throughout the range of

(21)

sediments tested. These results are consistent with the predictions based on the Dade et al. (1990) optimal foraging model and support the conclusion that A. pacifica alters behavior (i.e. changes feeding rate) in response to variations in food resources in such a way as to maximize its energy intake and thereby maximize fitness.

Acknowledgements

This research was supported by USEPA grant R823575-01. We thank Charlotte M. Fuller for instruction in the procedure for the measurement of tissue glycogen, Marilyn Mayer for instruction in sediment protein measurement, and various labmates for assisting in the collection of large volumes of sediment for the experiments. [RW]

References

Beukema, J.J., DeVlas, J., 1979. Population parameters of the lugworm, Arenicola marina, living on tidal flats in the Dutch Wadden Sea. Neth. J. Sea Res. 13, 331–353.

Bridges, T.S., 1996. Effects of organic additions to sediment, and maternal age and size, on patterns of offspring investment and performance in two opportunistic deposit-feeding polychaetes. Mar. Biol. 125, 345–357.

Bridges, T.S., Levin, L.A., Cabrera, D., Plaia, G., 1994. Effects of sediment amended with sewage, algae, or hydrocarbons on growth and reproduction in two opportunistic polychaetes. J. Exp. Mar. Biol. Ecol. 177, 99–119.

Cadee, G.C., 1976. Sediment reworking by Arenicola marina on tidal flats in the Dutch Wadden Sea. Neth. J. Sea Res. 10 (4), 440–460.

Calow, P., 1982. Homeostasis and fitness. Am. Nat. 120, 416–419.

Cammen, L.M., 1989. The relationship between ingestion rate of deposit feeders and sediment nutritional value. In: Lopez, G., Taghon, G., Levinton, J. (Eds.), Ecology of Marine Deposit Feeders. Springer-Verlag, New York, pp. 201–222.

Chernick, S.S., 1969. Determination of glycerol in acyl glycerols. Meth. Enzymol. 14, 627–630.

Chesney, Jr. E.J., Tenore, K.R., 1985. Oscillations of laboratory populations of the polychaete Capitella

capitata (Type I): their cause and implications for natural populations. Mar. Ecol. Prog. Ser. 20, 289–296.

Cleveland, W.S., 1979. Robust locally weighted regression and smoothing scatterplots. J. Am. Stat. Assn. 74 (368), 829–836.

Cleveland, W.S., 1981. LOWESS: A program for smoothing scatterplots by robust locally weighted regression. Am. Stat. 35, 54.

Dade, W.B., Jumars, P.A., Penry, D.L., 1990. Supply-side optimization: maximizing absorptive rates. In: Hughes, R.N. (Ed.), Behavioral Mechanisms of Food Selection. Springer-Verlag, Berlin, Heidelberg, pp. 531–556.

Farke, H., Berghuis, E.M., 1979. Spawning, larval development and migration behavior of Arenicola marina in the laboratory. Neth. J. Sea Res. 13, 512–528.

Forbes, T.L., Lopez, G.R., 1990. The effect of food concentration, body size, and environmental oxygen tension on the growth of the deposit-feeding polychaete Capitella species 1. Limnol. Oceanogr. 35, 1535–1544.

Giese, A.C., 1966. On the biochemical constitution of some echinoderms. In: Boolootian, R.A. (Ed.), Physiology of Echinoderms. Interscience Publishers, New York, pp. 757–796.

Gordon, D.C.J., 1966. The effects of the deposit feeding polychaete Pectinaria gouldii on the intertidal sediments of Barnstable Harbor. Limnol. Oceanogr. 11, 327–332.

(22)

Grassle, J.F., Grassle, J.P., 1974. Opportunistic life histories and genetic systems in marine benthic polychaetes. J. Mar. Res. 32 (2), 253–284.

Gremare, A., 1994. What describes fecundity of Capitella sp. 1 better: macro- or micronutrient availability? Mar. Biol. 119, 367–374.

Gremare, A., Marsh, A.G., Tenore, K.R., 1988. Short-term reproductive responses of Capitella sp. I (Annelida: Polychaeta) fed on different diets. J. Exp. Mar. Biol. Ecol. 123 (2), 147–162.

Gremare, A., Marsh, A.G., Tenore, K.R., 1989a. Fecundity and energy partitioning in Capitella capitata type I (Annelida: Polychaeta). Mar. Biol. 100 (3), 365–371.

Gremare, A., Marsh, A.G., Tenore, K.R., 1989b. Secondary production and reproduction of Capitella capitata type I (Annelida: Polychaeta) during a population cycle. Mar. Ecol. Prog. Ser. 51, 99–105.

Grime, J.P., 1977. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. Am. Nat. 111 (982), 1169–1194.

Healy, E., Wells, G.P., 1959. Three new lugworms (Arenicolidae: Polychaeta) from the North Pacific area. Proc. Zool. Soc. Lon. 133 (2), 315–335.

Hobson, K.D., 1967. The feeding and ecology of two North Pacific Abarenicola species (Arenicolidae: Polychaeta). Biol. Bull. 133 (2), 343–354.

Horng, C.-Y., 1998. Influence of the marine polychaete, Capitella sp. I, on the fate of sediment-bound polycyclic aromatic hydrocarbons-the role of feeding activity. Ph. D. Thesis, Rutgers, the State University of New Jersey.

Howie, D.I.D., 1961. The spawning of Arenicola marina (L.), III. Maturation and shedding of the ova. J. Mar. Biol. Assn. UK 41, 771–783.

Huitema, B.E., 1980. The Analysis of Covariance and Alternatives. John Wiley and Sons, New York. Hylleberg, J., 1975. Selective feeding by Abarenicola pacifica with notes on Abarenicola vagabunda and a

concept of gardening in lugworms. Ophelia 14, 113–137.

Karrh, R.R., Miller, D.C., 1994. Functional response of a surface-deposit feeder, Saccoglossus kowalevskii. Limnol. Oceanogr. 39 (6), 1455–1464.

Keppler, D., Decker, K., 1984. Glycogen. In: Bergmeyer, H.U. (Ed.), Methods of Enzymatic Analysis. Verlag Chemie, Weinheim, pp. 11–18.

Kudenov, J.D., 1982. Rates of seasonal sediment reworking in Axiothella rubrocinta (Polychaeta: Maldanidae). Mar. Biol. 70, 181–186.

Levin, L.A., Creed, E.L., 1986. Effect of temperature and food availability on reproductive responses of

Streblospio benedicti (Polychaeta: Spionidae) with planktotrophic or lecithotrophic development. Mar. Biol.

92, 103–113.

Longbottom, M.R., 1970. The distribution of Arenicola marina (L.) with particular reference to the effects of particle size and organic matter of the sediments. J. Exp. Mar. Biol. Ecol. 5, 138–157.

Lopez, G.R., Levinton, J.S., 1987. Ecology of deposit-feeding animals in marine sediments. Q. Rev. Biol. 62 (3), 235–260.

Marsh, A.G., Gremare, A., Tenore, K.R., 1989. Effect of food type and ration on growth of juvenile Capitella sp. I (Annelida: Polychaeta): macro- and micronutrients. Mar. Biol. 102 (4), 519–527.

Mayer, L.M., Schick, L.L., Sawyer, T., Plante, C.J., Jumars, P., Self, R.L., 1995. Bioavailable amino acids in sediments: A biomimetic, kinetics-based approach. Limnol. Oceanogr. 40 (3), 511–520.

Mayer, L.M., Schick, L.L., Setchell, F.W., 1986. Measurement of protein in nearshore marine sediments. Mar. Ecol. Prog. Ser. 30, 159–165.

McCall, P.L., 1977. Community patterns and adaptive strategies of the infaunal benthos of Long Island Sound. J. Mar. Res. 35, 221–266.

McGowan, M.W., Artiss, J.D., Strandbergh, D.R., Zak, B., 1983. A peroxidase-coupled method for the colorimetric determination of serum triglycerides. Clin. Chem. 29, 538.

Newell, G.E., 1948. A contribution to our knowledge of the life history of Arenicola marina L. J. Mar. Biol. Assn. UK 27, 554–580.

Nichols, F.H., 1974. Sediment turnover by a deposit-feeding polychaete. Limnol. Oceanogr. 19, 945–950. Pearson, T.H., Rosenberg, R., 1978. Macrobenthic succession in relation to organic enrichment and pollution

of the marine environment. Oceanogr. Mar. Biol. Annu. Rev. 16, 229–311.

(23)

Phillips, N.W., 1984. Compensatory intake can be consistent with an optimal foraging model. Am. Nat. 123, 867–872.

Prevedelli, D., Zunarelli Vandini, R., 1998. Effect of diet on reproductive characteristics of Ophryotrocha

labronica (Polychaeta: Dorvilleidae). Mar. Biol. 132, 163–170.

Radford, P.J., 1967. Growth analysis formulae-their use and abuse. Crop. Sci. 7 (3), 171–175. Radin, N.S., 1981. Extraction of tissue lipids with a solvent of low toxicity. Meth. Enzymol. 72, 5–7. Stearns, S.C., 1976. Life-history tactics: a review of the ideas. Q. Rev. Biol. 51 (1), 3–47.

Strathmann, M.F., 1987. Reproduction and Development of Marine Invertebrates On the Northern Pacific Coast. University of Washington Press, Seattle.

Taghon, G.L., 1981. Beyond selection: optimal ingestion rate as a function of food value. Am. Nat. 118 (2), 202–214.

Taghon, G.L., Greene, R.R., 1990. Effects of sediment-protein concentration on feeding and growth rates of

Abarenicola pacifica Healy et Wells (Polychaeta: Arenicolidae). J. Exp. Mar. Biol. Ecol. 136 (3), 197–216.

Taghon, G.L., Greene, R.R., Bard, D., 1990. Effects of food value of artificial and natural sediments on functional response and net rate of energy gain by a deposit-feeding polychaete. In: Hughes, R.N. (Ed.), Behavioral Mechanisms of Food Selection. Springer-Verlag, Berlin, Heidelberg, pp. 515–529.

Taghon, G.L., Jumars, P.A., 1984. Variable ingestion rate and its role in optimal foraging behavior of marine deposit feeders. Ecology 65 (2), 549–558.

Taghon, G.L., Prahl, F.G., Sparrow, M., Fuller, C.M., 1994. Lipid class and glycogen content of the lugworm

Abarenicola pacifica in relation to age, growth rate and reproductive condition. Mar. Biol. 120, 287–295.

Tenore, K.R., 1977. Growth of Capitella capitata cultured on various levels of detritus derived from different sources. Limnol. Oceanogr. 22 (5), 936–941.

Tenore, K.R., 1983. Organic nitrogen content and caloric content of detritus III. Effect on growth of a deposit-feeding polychaete, Capitella capitata. Estuarine Coastal Shelf Sci. 17, 733–742.

Tenore, K.R., Chesney, Jr. E.J., 1985. The effects of interaction of rate of food supply and population density on the bioenergetics of the opportunistic polychaete, Capitella capitata (type I). Limnol. Oceanogr. 30 (6), 1188–1195.

Thistle, D., 1981. Natural physical disturbances and communities of marine soft bottoms. Mar. Ecol. Prog. Ser. 6, 223–228.

Tsutsumi, H., 1990. Population persistence of Capitella sp. (Polychaeta: Capitellidae) on a mud flat subject to environmental disturbance by organic enrichment. Mar. Ecol. Prog. Ser. 63 (2 / 3), 147–156.

Tsutsumi, H., Fukunaga, S., Fujita, N., Sumida, M., 1990. Relationship between growth of Capitella sp. and organic enrichment of the sediment. Mar. Ecol. Prog. Ser. 63 (2 / 3), 157–162.

de Vooys, C.G.N., 1975. Glycogen and total lipids in the lugworm (Arenicola marina) in relation to reproduction. Neth. J. Sea Res. 9 (3–4), 311–319.

Whitlatch, R.B., Weinberg, J.R., 1982. Factors influencing particle selection and feeding rate in the polychaete

Cistenides(Pectinaria) gouldii. Mar. Biol. 71, 33–40.

de Wilde, P.A.W.J., Berghuis, E.M., 1979. Growth experiments on juvenile lugworms, Arenicola marina, in the laboratory. Neth. J. Sea Res. 13, 487–502.

Wilson, Jr. W.H., 1981. Sediment-mediated interactions in a densely populated infaunal assemblage: the effects of the polychaete Abarenicola pacifica. J. Mar. Res. 39 (4), 735–748.

Yingst, J.Y., 1982. Factors influencing rates of sediment ingestion by Parastichopus parvimensis (Clark), an epibenthic deposit-feeding holothurian. Estuarine Coastal Shelf Sci. 14, 119–134.

(1)

these treatments longer to reach these body sizes, but they were feeding at higher rates than worms in the richer treatments were when they were at that same size. Some of the different response patterns seen among worms feeding in sediments with different protein organic concentrations may be due to the differential timing of allometric shifts in feeding, growth, and progression of reproduction. From the perspective of comparing these data with model predictions and previous experimental results, these different trajectories would lead to different conclusions depending on the size / age and prior nutritional history of the worms used in separate experiments. Small (and presumably young) worms fed faster in sediments with higher organic concentrations. Larger animals (75–100 mg AFDW) showed the opposite pattern. There are no reliable methods for aging soft-bodied marine invertebrates such as deposit-feeding polychaetes. The usual assumptions are that similarly sized individuals are of similar age and have similar nutritional histories. Errors in these assumptions could lead to opposite conclusions when comparing experiments, even experiments using the same species. Ideally, the age and nutritional history of animals used in experiments should be known, but this is not always practical. It is often much easier to use animals that were recently collected from the field in laboratory experiments. Doing so may introduce unexpected levels of variability, however, making interpretation of results more ambiguous. Due to these complexities, we must concentrate our analysis on differences among treatments within this experiment. The pattern of higher feeding rates in sediments with greater organic concentration is not inconsistent with the optimal foraging model of Dade et al. (1990) for sediments below the ‘threshold’ organic concentration.

4.2. Growth

|100 mg AFDW (Fig. 4B). Growth rates at this size were ¯1% / day. Nevertheless,

worms in the higher quality sediments were always significantly larger and worms in all treatments continued to grow, even while completing vitellogenesis.

(2)

Capitella sp. I (Marsh et al., 1989; Gremare, 1994). Tsutsumi et al. (1990) found that Capitella sp. I grew at significantly different rates in two sediments with the same

protein concentration (|4 mg / g) in two different experiments. In the experiment with

higher growth rates the sediment had been enriched to this level by addition of powdered algae, while the other used native sediment with this protein concentration. The sources of protein in the two experiments were different and worms responded to them differently. Hylleberg (1975) found that while ciliates, flagellates, nematodes, motile bacteria, and some motile diatoms were readily digested by Abarenicola pacifica, bacteria and diatoms attached to sediment grains and detrital fragments passed the gut undigested. All of these sources might be measured as bioavailable protein, but not all may actually be available to deposit feeders. Therefore, different sediments, with their complex mixes of these components, may have the same protein concentration but be of different value to organisms and elicit different functional and physiological responses.

4.3. Reproduction

(3)

reproductive cycle (de Vooys, 1975). Tissue protein concentrations also showed no differences between treatments.

Abarenicola pacifica.

´ ´

Capitella sp. I (Gremare et al., 1988, 1989a; Bridges et al., 1994; Gremare, 1994).

Increases in fecundity have also been seen in Capitella sp. I in response to increased ´

labronica had no change in egg size related to different diets (Prevedelli and Zunarelli

Vandini, 1998). Similarly, in the present study, egg sizes of Abarenicola pacifica were not significantly different across a range of sediment quality.