@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Science, Faculty of"@en, "Zoology, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Hollett, Lois Julia"@en ; dcterms:issued "2010-08-16T20:50:59Z"@en, "1988"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description "The presence of natural and artificial eelgrass reduced predation by the Dungeness crab Cancer magister on three prey species: a bivalve, Macoma nasuta, a shrimp, Crangon franciscorum, and a fish, Gasterosteus aculeatus. Overall, when provided with a choice of the three prey species, crabs foraging in an eelgrass habitat consumed less total prey biomass than crabs preying in a bare habitat. The three species varied in their susceptibility to predation by Cancer magister and this variation generally reflected the microhabitat occupied by them. The epifaunal shrimp appeared to be most susceptible to predation, followed by the more motile fish and, finally, the infaunal clam."@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/27476?expand=metadata"@en ; skos:note "The Effect Of Structural Complexity In Eelgrass Meadows On The Predatory Activity Of The Dungeness Crab Cancer Magister by Lois Julia Hollett A Thesis Submitted in Partial Fufflllment of The Requirements for the Degree of Master of Science in The Faculty of Graduate Studies (Zoology) We accept this thesis as conforming to the required standard The University of British Columbia December 1988 © Lois Julia Hollett 19 8 8 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada DE-6 (2/88) ii ABSTRACT The presence of natural and artificial eelgrass reduced predation by the Dungeness crab Cancer magister on three prey species: a bivalve, Macoma nasuta, a shrimp, Cran-gon franciscorum, and a fish, Gasterosteus aculeatus. Overall, when provided with a choice of the three prey species, crabs foraging in an eelgrass habitat consumed less total prey biomass than crabs preying in a bare habitat. The three species varied in their susceptibility to predation by Cancer magister and this variation generally reflected the microhabitat occupied by them. The epifaunal shrimp appeared to be most susceptible to predation, followed by the more motile fish and, finally, the infaunal clam. iii Table of Contents Page Abstract \" List of Tables . . iv List of Figures v Acknowledgements vi Introduction 1 Methods Field Site Description 4 Species Description Predator . 4 Prey 7 Field and Laboratory Enclosures Field Enclosures 9 Laboratory Enclosures 11 Experimental Procedure 13 Single-Prey Experiments . . 14 Multiple-Prey Experiments 14 Statistical Analysis 15 Results 16 Single-Prey Experiments 19 Multiple-Prey Experiments • • • 23 Discussion 30 References . 36 iv List of Tables Page 1. Shoot density of the single- and multiple-prey field experiments and a summary of the ANOVA statistics comparing eelgrass shoot densities of the field experiments 17 2. Size of crabs utilized in the field and laboratory single- and multiple-prey experiments and a summary of the ANOVA statistics comparing mean size of crab . 18 3. Summary of the ANOVA statistics comparing the number of individual prey consumed in the bare enclosures with that in the eelgrass enclosures for the field single-prey experiments 21 4. Summary of the repeated-measures ANOVA statistics comparing the number of individual prey consumed in the bare enclosures with that of the eelgrass enclosures for the laboratory single-prey experiment 21 5. Summary of the repeated-measures ANOVA statistics comparing the number of individual prey consumed in the bare enclosures with that of the eelgrass enclosures for the field multiple-prey experiment 25 6. Summary of the repeated-measures ANOVA statistics comparing the number of individual prey consumed in the bare enclosure with that of the eelgrass enclosures for the laboratory multiple-prey experiment 25 7. Summary of the repeated-measures ANOVA statistics comparing the gram biomass consumed in the bare enclosures with that of the eelgrass enclosures for the field and laboratory multiple-prey experiments 29 V List of Figures Page 1. Map showing the location of the field study site within the Roberts Bank seagrass meadow 5 2. Schematic diagram of a field enclosure and skirt 10 3. Schematic diagram of laboratory enclosures 12 4. Number of individual Macoma nasuta, Crangon franciscorum, and Gasterosteus aculeatus consumed per crab in the bare and eelgrass enclosures of the field single-prey experiments 20 5. Number of individual Macoma nasuta, Crangon franciscorum, and Gasterosteus aculeatus consumed per crab over 72 hours in the bare and eelgrass enclosures of the laboratory single-prey experiments. . . 22 6. Number of individual Macoma nasuta, Crangon franciscorum, and Gasterosteus aculeatus consumed per crab in the bare and eelgrass enclosures of the 24, 48, 72, and the 144 hour field multiple-prey ex-periments 24 7. Number of individual Macoma nasuta, Crangon franciscorum, and Gasterosteus aculeatus consumed per crab over 144 hours in the bare and eelgrass enclosures of the laboratory multiple-prey experiment . 27 8. Biomass consumed per crab in the bare and eelgrass enclosures of the field and laboratory multiple-prey experiments 28 vi Acknowledgements I would like to thank my supervisor, Tom Carefoot, for guiding me through my exploration of the marine ecosystem. And to my committee members, Bill Neill and Paul G. Harrison, my sincere gratitude for many hours of conversation and for their in-sightful guidance throughout this study. The members of Paul G. Harrison's lab, Cindy, Farida, Kafhy, Eric, and Pat, made the many days that we spent at Roberts Bank very pleasant and their enthusiasm about eelgrass ecology provided the motivation for much of my research. And finally, I would like to thank my friend, Locke Rowe, for accom-panying me on many late night sampling quests and for his tireless support through this and many other adventures. 1 INTRODUCTION A seagrass meadow is one of the most structurally complex of all estuarine habitats. Structural complexity in this habitat is provided by the physical presence of vegetative shoots of seagrass in the water column and by a dense rhizome mat in the substrate. Fauna associated with this complexity are characterized by high species abundance (Orth, 1977; Reise, 1978; Stoner, 1980; Lewis and Stoner, 1983; Summerson and Peterson, 1984; Posey, 1988) and biomass (Hall and Bell, 1988). While this rich faunal as-semblage probably results from a combination of factors such as higher larval settlement rates (Eckman, 1984, 1987), abundant food (Peterson et al. 1984) and suitable substrate (Stoner, 1980; Peterson and Black, 1986), predation is regarded as a key force in deter-mining the composition of seagrass communities (Orth, 1977; Young and Young, 1977; Virnstein, 1978; Nelson, 1979a, 1981; Lewis, 1984; Summerson and Peterson, 1984; Virnstein et al. 1984; Leber, 1985). The structural complexity of seagrass meadows reduces the rate of predation by providing a refuge for various prey species (Reise, 1978; Nelson, 1979a; Coen et al. 1981; Heck and Thoman, 1981; Stoner, 1982; Leber, 1985; Main, 1987). However, the precise nature of the relationship between structural complexity of seagrass meadows and predation risk appears to depend upon the specific predator and prey species in ques-tion. For example, the degree to which predator foraging efficiency may be reduced in these habitats will depend upon that predator's foraging tactics; an actively searching predator will be inhibited more than will an ambush predator (Coen et al. 1981; Nel-son, 1981). Some variables which may contribute to a prey's success at obtaining refuge within such a complex habitat include prey size (Arnold, 1984; Jensen and Jensen, 1985), cryptic coloration (Main, 1987), and characteristics of the microhabitat occupied, 2 such as the presence of seagrass leaves or rhizomes providing a refuge for epifauna and infauna respectively (Young et al. 1976; Nelson, 1981; Lewis and Stoner, 1983; Ar-nold, 1984; Smith and Coull, 1987). Researchers have also documented an influence of indirect variables such as season (Holland et al. 1980; Heck and Thoman, 1981), diurnal period (Greening and Livingston, 1982; Summerson and Peterson, 1984), and competition between prey (Coen et al. 1981) on the interaction between habitat com-plexity and predation. Given this species-specific variation in these several potentially influencing fac-tors, the outcome of predator-prey interactions in seagrass meadows is often difficult to predict, and the measures such as total species abundance and biomass may actually mask the different responses of component species (Bell and Westoby, 1986). An in-vestigation of the effect of seagrass on predation should, therefore, include detailed studies of predator-prey interactions at the species level. The seagrasses Zostera marina L. and Z. japonica Aschers. and Graebn. form con-spicuous seagrass meadows in protected estuaries of the Pacific Northwest. The largest of the benthic predators in this system is the Dungeness crab Cancer magister Dana. The principal benefits to crabs living in a seagrass meadow appear to be an abundant food source and warmer temperatures, leading to increased growth rates relative to off-shore populations (Stevens and Armstrong, 1984; Armstrong and Gunderson, 1985), and perhaps a refuge from their own predators (Botsford and Wickham, 1978; Stevens et al. 1982). Cancer magister is an opportunistic carnivore (omnivore), preying upon a broad spectrum of infaunal, epifaunal, and neritic phyla (Butler, 1954; Gotshall, 1977; Stevens et al. 1982). Given this broad diet, the presence of seagrass may alter predation rates on all prey types or it may bias predation to a few prey species. A bias in prey selec-tion might arise from one or more of the potential prey species having a higher suscep-3 tibility to predation than other potential prey species. In the present study, I investigated the effect of structural complexity (seagrass) on predation by Cancer magister of three prey species, the bivalve Macoma nasuta Con-rad, the shrimp Crangon franciscorum Stimpson, and the fish Gasterosteus aculeatus L. Field and laboratory enclosure experiments compared the number of prey consumed in enclosures containing natural or artificial Zostera marina to the number of prey con-sumed in enclosures containing bare substrate. Two types of experiments were per-formed: one involving a single prey species to determine the susceptibility of the in-dividual prey species to predation in the two habitats, and one involving all three prey to examine the potential interaction between the three prey species. The general predic-tion for all experiments was that seagrass reduces predatory activity of the crab and thereby provides protection to all three prey species. 4 METHODS Field Site Description All field surveys and experiments were carried out in the intercauseway seagrass meadow at Roberts Bank, British Columbia (42°2'N; 23°8'W; Fig. 1). This exten-sive meadow of approximately 430 ha is bounded on the northwest by the Roberts Bank coalport causeway and by the Tsawwassen ferry terminal causeway to the southeast. This meadow consists of 2 species of seagrass: Zoster a japonica dominates the upper tidal region of the seagrass zone while eelgrass Z. marina dominates the lower tidal region. All field experiments took place in the region of Z. marina. This species is a perennial seagrass that maintains leafy shoots all year, ranging in length from 25-75 cm and in density from 25-150 vegetative shoots.m . This estuary has a mixed maritime climate with cool wet winters and warm dry summers. During the period of field experiments (April - August, 1988), the lowest of the two daily low tides occurred from mid-morning to early afternoon. Species Descriptions Predator: Cancer magister The Dungeness crab Cancer magister ranges from the Aleutian Islands to northern Mexico (Hart, 1982) and is commercially exploited throughout most of its range (Jamieson, 1986). C. magister are common in coastal near-shore and estuarine habitats. Within an estuary, seagrass meadows appear to be an important habitat for the post-lar-val and juvenile crabs (Butler, 1956; Lough, 1976; Stevens and Armstrong, 1984; 5 Figure 1. A) Location of the intercauseway area in the Fraser River estuary, British Columbia. B) Map of Roberts Bank eelgrass meadow (Stippled area: region of Zostera japonica; shaded area: region of Z. marina; clear area: region of bare sand and/or mud substrate; solid circles: navigation towers; solid line: location of transect for June, 1987 crab survey; X: location of field enclosures). 6 Armstrong and Gunderson, 1985). Bivalves, crustaceans and juvenile fish are reported to be the most important prey types of Cancer magister, while the specific prey species and order of importance as prey vary with geographic location, time of day, and season (Butler, 1954; Gotshall, 1977; Stevens et al, 1982). In addition, an ontogenetic switch in food preference is probably a general phenomenon for the Dungeness crab. For example, in Grays Har-bor, Washington, first-year crabs consumed a relatively high proportion of small bival-ves (Cryptomya californica, Macoma sp., and Tellina sp.), while shrimp (Crangon sp.) were consumed in a relatively high proportion by second-year crabs, and third-year crabs consumed high proportions of juvenile teleost fish such as sandlance (Ammodytes hexap-tera), sanddab (Citharichthys sordidus), lingcod (Ophiodon elongatus), and shiner perch (Cymatogaster aggregata; Stevens et al., 1982). This ontogenetic shift in diet may reflect age or a size-related prey handling ability in conjunction with the distribution and abundance of prey. A preliminary survey conducted during the period of the lowest tide in June 1987 provided a rough estimate of crab distribution and density at the Roberts Bank study site. This was done by establishing a transect 200 m long running seaward from an area of patchily vegetated Zostera japonica and Z. marina into a pure stand of Z. marina. At 5 points along the transect, approximately 50 m apart, I manually searched three 9-m quadrats for crabs. These three quadrats were 5 m apart, running perpendicular to the transect with the middle quadrat being positioned exactly on the transect. The num-ber, size, and sex of crabs from each quadrat were recorded, as well as seagrass shoot density. The density of Z. marina ranged from 60-116 shoots'm with a mean density of 89.6±18.5 (S.E.) shoots'm ' A total of 81 crabs were found, ranging in size from 21.7-58.8 mm (maximum carapace width) with a mean size of 34.5+1.6 (S.E.) mm. The density of crabs ranged from 0.2-2.0 crabs.m\" for a mean density of 0.7±0.4 (S.E.) crabs'm\"2. 7 Crabs for all experiments were collected from Roberts Bank. They were main-tained in the flow-through seawater system at the Bioscience Building of the Univer-sity of British Columbia and fed ad libitum on crushed mussels Mytilus edulis. Prior to each field and laboratory experiment, crabs were separated into individual flow-through containers and fed a crushed mussel. Those crabs which did not eat within 24 hours were not used in the experiment. Those which did, were then starved for 24 hours in order to standardize their gut condition. The purpose of this procedure was to decrease the variation in crab response to the experimental conditions. At the end of an experi-ment crabs were returned to the laboratory stock and a random sample of this stock was then prepared for the next experiment. Due to the extensive time required to collect crabs from Roberts Bank, crabs were reused in field and laboratory experiments. However, I ensured that they were only used in alternate experiments. A total of 53 individual crabs was used in the study. Prey: Three prey species were employed, the bent-nosed clam Macoma nasuta, the bay shrimp Crangon franciscorum, and the three-spine stickleback Gasterosteus aculeatus. Their selection was based initially on the importance, in broad terms, of bivalves, crus-taceans and fish in the diet of Cancer magister (Butler, 1954; Gotshall, 1977; Stevens et al., 1982). All are common, year-round residents of seagrass meadows (Phillips, 1984). In addition, these three prey species are known to occupy the three main prey microhabitats exploited by the crabs: infauna (clams), epifauna (shrimp), and neritic (fish). They provide a comparison of susceptibility to predation in the habitats com-pared in the present study. In addition to measuring consumption of a certain prey species by the crab, it was necessary to have a standard measure of their biomass which could be used as a com-8 mon measure for all three prey species. From length/wet weight relationships calculated for each prey species, the mean length measured for each prey type was converted into a measure of biomass in grams. This measure was then used to convert the number of individuals consumed into grams of biomass consumed. The bent-nosed clam Macoma nasuta is a common infaunal bivalve inhabiting quiet-water bays from Kodiak Island, Alaska to Cabo San Lucas, California (Coan, 1971). It lives buried to approximately 10-15 cm in sand or mud substrates. M. nasuta were col-lected from the Roberts Bank field site from the sediment layer just below the eelgrass 2 2 rhizome mat. Densities ranged from 17-32'm\" (mean density of 24+1 (S.E.) 'm , N=14). The M. nasuta used in lab and field experiments ranged from 24-56 mm in total length (mean length of 38.6±0.7 (S.E.) mm, N=100). From the length/weight relationship of Wt=0.0006.L2,266 (N=20,1^ =0.919), the mean biomass of an individual M. nasuta (excluding the shell) is 2.36 g. Crangon franciscorum is a euryhaline caridean shrimp which ranges from Resur-rection Bay, Alaska to San Diego, California (Butler, 1980). Shrimp were collected from Spanish Banks in Burrard Inlet, British Columbia, using a seine net. This method of collection made it impossible to get more than a rough estimate of their natural den-sity (reported densities of Crangon sp. vary geographically, seasonally, and diurnally; Stevens et al., 1982; Gee et al., 1985). The shrimp used in field and laboratory ex-periments ranged from 4-14 mm in carapace length (mean of 8.7±0.2 (S.E.) mm, N=100). From the length/weight relationship of Wt=0.002L273 (N=20, r2=0.988), the mean biomass of an individual C. franciscorum is 0.73 g. The three-spine stickleback Gasterosteus aculeatus ranges in marine habitats from Aleutian Island, Alaska to Baja, California (Hart, 1973). They were collected in min-now traps from freshwater drainage ditches in the southlands area of Vancouver, British Columbia and, in preparation for the field and laboratory experiments, were gradually acclimated to seawater over a 48 hour period. The stickleback used in field and 9 laboratory experiments ranged from 33-65 mm in total length (mean of 44.5±0.7 (S.E.) mm, N=100). From the length/weight relationship of Wt=0.000L3,284 (N=20, r2=0.954), the mean biomass of an individual G. aculeatus is 0.65 g. All three prey species were maintained in the flow-through seawater system at the University of British Columbia and fed ad libitum on fish chow prior to being used in the experiments. Field and Laboratory Enclosures Field Enclosures: Field enclosures consisted of closed, cylindrical cages measuring 1 m in diameter and 0.5 m in height, enclosing an area of 0.79 m . The enclosures consisted of an aluminum frame covered on the side and top surfaces by a layer of 7 mm aluminum wire mesh and on the outside by a layer of 4 mm vexar mesh (Fig. 2). Each enclosure was surrounded by a sheet metal skirt, 0.2 m in height. The skirts were pushed into the sediment approximately 15 cm, leaving 5 cm above the sediment within which the lower edges of the enclosures were placed. The skirts functioned in three ways: 1) they restricted the growth of eelgrass within the enclosures by inhibiting the spread of rhizomes, 2) they served to enclose Macoma nasuta as well as impede the escape of crabs, and 3) they anchored the enclosures. Six enclosures were erected in the area where the June 1987 field survey indicated a high density of Zostera marina and Cancer magister (Fig. 1). The enclosures were placed 5 m apart in a line perpendicular to the movement of the tides. At the beginning of each experiment the sheet metal skirts were positioned at the experimental site. Within three of the skirts, the eelgrass shoots and associated rhizomes were cut and removed leaving only the bare sediment. This disturbance did not appear Figure 2. Schematic diagram of a field enclosure and skirt 11 to change the nature of the sediments (i.e. compaction) within these enclosures. The other three retained the natural growth of eelgrass shoots and rhizomes. Visible mac-rofauna (fish, crabs and bivalves) were removed from all enclosures and the six enclosures were then placed within the skirts. The design of the field experiments consisted of two control enclosures, one bare and one eelgrass, which contained only prey, and four experimental enclosures, two bare and two eelgrass, which contained both crabs and prey. Laboratory Enclosures: Laboratory enclosures were constructed in 3 concrete tanks (1.70 m long x 0.80 m wide x 0.25 m deep) each with a plywood bottom. Each tank was separated into 4 compartments (0.43 m long x 0.8 m wide x 0.25 m deep, area = 0.34 m ) using wooden frame dividers covered with 4mm vexar mesh (Fig. 3). The 12 resulting enclosures were provided with a minimum of 5 cm depth of silica sand. A constant water flow (1.7 L'min\"1) was delivered to each enclosure, emptying into a central drain. The water temperature and salinity were maintained at approximately 10°C and 28 °/oo respective-ly. Two enclosures in each tank were left with only the sand substrate to simulate the bare field enclosures, while the other two contained artificial eelgrass. Strips, each measuring 75.0 cm long x 0.75 cm wide, cut from polyplastic \"solar pool blanket\" were used to simulate eelgrass shoots. Four strips were stapled together at one end to form a single \"shoot\". Each eelgrass enclosure contained 52 such shoots stapled onto the plywood bottom (density=150 shoots'm ). In addition, when Macoma nasuta were used in an experiment, natural rhizomes were embedded in the sand substrate of each \"eelgrass\" enclosure to simulate below-ground structural complexity. Since these laboratory experiments were relatively short-term (maximum 144 hours), natural rhizomes could be used (these were discarded after each experiment). Rhizomes were 12 0.81m Figure 3. Schematic diagram of laboratory enclosures. The plan view shows one of 3 concrete tanks subdivided into 4 compartments. 13 dug from the Roberts Bank field site, cleaned of excess sediment and infauna, and were dug into the sand substrate of each artificial eelgrass enclosure. The quantity of rhizomes used was a wet measure corresponding to an estimate of rhizome wet weight:eelgrass 2 shoot density in the field enclosures (710 g rhizomes: 150 eelgrass shoots'm , N=9). The design of the laboratory experiments was three replicates of four treatments. The treatments consisted of two control enclosures (prey only), one bare and one artifi-cial eelgrass, and two experimental enclosures (predator and prey), one bare and one ar-tificial eelgrass, each replicated 3 times for a total of 12 enclosures. Experimental Procedure Field experiments were conducted from April - August 1988 and laboratory experi-ments from February - August 1988. Once the field and laboratory enclosures were es-tablished, a known number of prey was introduced to all enclosures. The prey were given sufficient time to burrow or disperse before introducing the predator. This set-tling-in time ranged from approximately one hour for the fish and shrimp, to 24 hours for the bivalve. Prey species were either enclosed individually for single-prey experi-ments, or collectively for multiple-prey experiments. The density of prey varied some-what between these experiments and is outlined in a later section. Predator density within field and laboratory, single- and multiple-prey experiments was constant. Each 2 field experimental enclosure contained a total of 6 crabs (density of 7.6 crab'm ) and 2 each laboratory enclosure contained a total of 3 crabs (density of 8.8 crab'm ). At the termination of each field experiment, eelgrass shoots within the eelgrass enclosures were counted for a measure of shoot density. To ease in the capture of remaining prey, these enclosures were then cleared of eelgrass shoots and rhizomes. All crabs and remaining prey were then removed from all enclosures and counted. The 14 enclosures were then moved to a new site for the next experiment. In the laboratory experiments, counts were made of the surviving shrimp and/or fish while the number of consumed clams was measured by counting empty shells. Two types of experiments were conducted in both the field and the laboratory. Single-prey experiments determined the susceptibility of each prey species to predation in the 2 habitats while multiple-prey experiments examined how potential interactions between the three prey species may alter their individual or collective susceptibility to predation. In total, 7 field and 4 laboratory experiments were conducted. Single-Prey Experiments: One single-prey experiment was conducted for each prey type in both the field (24 hours in duration) and the laboratory (72 hours in duration). The number of prey con-sumed in the field experiments was counted at an end of the experiment while prey con-sumed in the laboratory experiments were monitored at least every 24 hours and more frequently for the first 48 hours. Similar prey densities of the three prey species were used for the field and laboratory _o experiments. Fifty prey were used in the field experiments yielding a density of 63'm 2 while 25 prey were used in the laboratory experiments, yielding a density of 73.5'm . This provided an initial 8.3 prey for each individual crab in both the field and laboratory experiments. Multiple-Prey Experiments: Four multiple-prey experiments were run in the field, varying in durations of 24 hours, 48 hours, 72 hours, and 144 hours. One multiple-prey experiment was run in the laboratory, 144 hours in duration. The number of prey consumed in the field ex-15 periment was determined at the end of the experiments while the number of prey con-sumed in the laboratory experiments was monitored at least every 24 hours and more frequently for the first 48 hours. Similar prey densities were used for field and laboratory experiments. In the field experiments, each enclosure contained a total of 75 individuals (25 of each species) 2 yielding a density of 95 individuals'm\" . Each laboratory enclosure contained a total of 45 individuals (15 of each species) for a density of 132 individuals'm . This provided each crab with an initial 4.2 individuals of each prey species in the field experiments and an initial 5.0 individuals of each prey species in the laboratory experiments. STATISTICAL ANALYSIS: A one-way analysis of variance was used to determine if field enclosures differed significantly in the enclosed density of natural Zoster a marina and similarly, to deter-mine if the mean size of crabs employed differed significantly within or between field and laboratory experiments. The difference between the number of prey still alive in the control enclosures and the number of prey still alive in the experimental enclosures was a measure of the num-ber of prey consumed in an experiment. All results are presented as a mean + stand-ard error. To examine the effects of structural complexity on crab predation, data from the field single-prey experiment were analyzed with a one-way ANOVA. Two-way, repeated-measures ANOVAs were used to test for the effect of natural and artificial eelgrass on prey consumption in the laboratory single-prey experiments and in the field and laboratory multiple-prey experiments as the data were collected continuously over a period of time. 16 RESULTS Shoot density in the eelgrass enclosures did not vary significantly within or be-tween field experiments (ANOVA, F=2.23, p=0.10, Table la&b). Mean shoot density of eelgrass enclosures for all field experiments was 126+3 shoots'm\" . This natural shoot density is somewhat lower than that of the laboratory experiments which was set at 150 2 ' artificial shoots'm . Crab size did not vary significantly within or between field and/or laboratory ex-periments (ANOVA; F=1.03, p=0.43, Table 2a&b). Crabs did not moult in any of the field or laboratory enclosures. Also, while there was no mortality of crabs in the laboratory experiments, 1 crab died in 4 of the 7 field experiments representing only a 2.4% mortality of a total of 168 crabs employed in the field experiments. The death of these crabs may have been due largely to the stress of being moved from the laboratory to the field, since 3 of the 4 corpses showed no visible sign of injury. The remainder of the crabs appeared to be in good health as they were very active immediately upon release into the enclosures and the results illustrate prey consumption within 24 hours of the beginning of the experiment. Recovery of prey from the control field enclosures ranged from 94-100% of the in-itial number enclosed. In the laboratory controls, only Crangon franciscorum declined in number. However, this mortality never amounted to more than 8% of the initial num-ber enclosed (a total of 300 in all single-prey experiments and a total of 180 in all mul-tiple-prey experiments), and generally occurred in the final 24 hours of an experiment. It appeared to be the result of cannibalism. 1 7 Table la. Shoot densities of the single- and multiple-prey field experiments (x±S.E.). E x p e r i m e n t S h o o t D e n s i t y ( x + S . E . ) Macoma nasuta 130.3+4.1 Single-Prey Crangon franciscorum 114.0+7.6 Gasterosteus aculeatus 112.0+10.3 24 Hours 131.0+7.5 Multiple-Prey 48 Hours 127.3+4.1 72 Hours 125.2+8.0 144 Hours 145.0+11.7 Table lb. Summary of the A N O V A statistics comparing eelgrass shoot densities in field experiments. S o u r c e o f V a r i a t i o n D F MS B e t w e e n E x p e r i m e n t s 6 2237.14 W i t h i n E x p e r i m e n t s 14 2336.00 2.23 0.10 18 Table 2a. Size of crab utilized in field and laboratory single- and multiple-prey experiments (x+S.E.). Experiment Treatment Crab Size (x+S.E.) Single-Prey Lab Multiple-Prey Lab M. nasuta Bare 102 .0+1 .5 Eelgrass 100 .6+0 .5 franciscorum Bare 100 .3+1 .9 Eelgrass 102 .2 + 1 .6 aculeatus Bare 99 .4+1 .1 Eelgrass 100 .1+2 .6 M. nasuta Bare 90 .5+3 .2 Eelgrass 88 .8+3 .4 franciscorum Bare 93 .8+2 .5 Eelgrass 95 .2+2 .7 aculeatus Bare 88 .4+2 .9 Eelgrass 88 .8+2 .8 24 Hours Bare 104 .6+0 . 6 Eelgrass 102 .7+2 .5 48 Hours Bare 103 .3+0 .7 Eelgrass 102 .8+0 .2 72 Hours Bare 96 .0+3 .6 Eelgrass 102 .7+2 .5 144 Hours Bare 102 . 9+2 .2 Eelgrass 101 .4 + 3 . 6 144 Hours Bare 101 .3 + 1 .6 Eelgrass 98 .7 + 1 .7 Table 2b. Summary of the ANOVA statistics comparing the mean size of crabs in field and laboratory experiments. Source of Variation DF MS F p Between Experiments 51 7172.73 1.03 0.43 Within Experiments 184 25159.70 19 Single-Prey Experiments: The effect of the presence eelgrass on predation of a single prey species by Can-cer magister varied with the prey and between field and laboratory experiments. In the field, crabs in the bare enclosures consumed significantly more Macoma nasuta than crabs in the eelgrass enclosures (ANOVA, F=18.77, p=0.037, Fig. 4, Table 3). In the laboratory experiment, however, there was no difference in the number of M. nasuta consumed between the two habitats (ANOVA, F=1.07, p=0.28, Fig. 5A, Table 4). Crabs had little difficulty locating and consuming the bivalves in the laboratory enclosures and approximately 80% of the initial prey number were consumed within 24 hours. The planted rhizome mat did not appear to be an obstruction and it was often dug up with the clams. Significantly more Crangon franciscorum were consumed in the field eelgrass enclosures relative to the bare enclosures (ANOVA, F=30.25, p=0.023, Fig. 4, Table 3). In contrast, the laboratory experiment showed that crabs in the bare enclosures consumed significantly more C. franciscorum than did the crabs in the eelgrass enclosures (ANOVA, F=51.20, p=0.0005, Fig. 5B, Table 4). In the laboratory experiment, the majority of shrimp were consumed in the bare enclosures within 48 hours. At this point each crab had consumed approximately 6 shrimp or 4.2 g of biomass, representing 72% of the starting biomass. The consumption of shrimp in the eelgrass enclosures continued more evenly throughout the 72 hour experiment after which time each crab had con-sumed only 2.4 shrimp or 1.7 g of biomass (29% of the starting biomass). In the field experiment (see Fig. 4), crabs consumed less than 2% of the available Gasterosteus aculeatus in both the bare and eelgrass enclosures and, thus, consumption did not differ significantly between the 2 habitats (ANOVA, F=1.92, p=0.29, see also Table 3). Crabs in the laboratory experiment, on the other hand, consumed up to 68% of the available stickleback (Figure 5C). Here, the majority of stickleback consumed in 20 a Macoma Crangon Gasterosteus nflSUta franciscorum aculeatus Figure 4. Number of individual Macoma nasuta, Crangon franciscorum, and Gasterosteus aculeatus consumed per crab over 24 hours in the bare (light stippled bars) and eelgrass (dark stippled bars) enclosures of the field single-prey experi-ments. (x±S.E., N=2) 21 Table 3. Summary of the ANOVA statistics comparing the number of individual prey consumed in the bare enclosures with that in the eelgrass enclosures for the field single-prey experiments. S o u r c e o f V a r i a t i o n D F MS M a c o m a n a s u t a Between Enclosures 1 Within Enclosures 2 C r a n g o n f r a n c i s c o r u m Between Enclosures 1 Within Enclosures 2 G a s t e r o s t e u s a c u l e a t u s Between Enclosures 1 Within Enclosures 2 42.25 4.50 121.00 4.00 25.00 13.00 18.77 0.037 30.25 0.02.3 1.92 0.293 Table 4. Summary of the repeated-measures ANOVA statistics comparing the num-ber of individual prey consumed in the bare enclosures with that of the eelgrass enclosures for the laboratory single-prey experiment. S o u r c e o f V a r i a t i o n D F MS F p F i e l d M u l t i p l e - P r e y E x p e r i m e n t s Time 3 24.34 6.13 0.09 Treatment 1 59.33 75.10 0.0005 Interaction 3 5.03 5.03 0.032 L a b o r a t o r y M u l t i p l e - P r e y E x p e r i m e n t Time 10 75.67 19.71 0.0005 Treatment 1 224.00 29.47 0.0005 Interaction 10 3.84 0.51 0.54 22 -I — i — 3 6 —T— 48 — i — 60 72 Time (Hours) Figure 5. Number of individual Macoma nasuta, Crangon franciscorum, and Gasterosteus aculeatus consumed per crab over 72 hours in the bare (solid line) and eelgrass (dashed line) enclosures of the laboratory single-prey experiments (x±S.E.,N=3). 23 the bare enclosures occurred within 12 hours of the start of the experiment, and there was no further consumption after 24 hours. At this time each crab had consumed 4.6 fish or 3.0 g of biomass (representing 55% of the starting biomass). In comparison, predation on fish in the eelgrass enclosure continued throughout the 72-hour experiment. There were significantly more fish consumed in the bare enclosures than in the eelgrass enclosures (ANOVA, F=78,13, p=0.0005, Table 4). In the laboratory single-prey experiments, the consumption of shrimp in the bare enclosures and fish in both enclosures appeared to be a declining function over time. This is unlikely to be the result of satiation. The experiment with Macoma nasuta demonstrated that crabs are able to consume up to 16.3 g of biomass in 24 hours and up to 19.2 g in 72 hours. After 72 hours, only 23% of this potential intake had oc-curred in the bare enclosures of the Crangon franciscorum experiment and only 21% in the bare enclosures of the Gasterosteus aculeatus experiment. In addition, although the number of prey consumed declined for the last half of these experiments, I observed that crabs continued to forage throughout all experiments. Therefore, this decrease in prey capture must be attributed to other factors, such as a learning response in the prey to avoid predation by the crab. Multiple-Prey Experiments: The consumption of individual prey species in the field multiple-prey experiment was similar to the results of the field single-prey experiments in terms of the relative number consumed (Fig. 6). Early in the experiment, Crangon franciscorum were con-sumed in large quantities while fewer Macoma nasuta and Gasterosteus aculeatus were consumed (Fig. 6). No clams were consumed in the eelgrass enclosure until after 72 hours (Fig. 6A). The number of clams consumed was significantly greater in the bare enclosures than in the eelgrass enclosures (ANOVA, F=44.89, p=0.0005, Table 5). 24 Figure 6. Number of individual Macoma nasuta, Crangon franciscorum, and Gasterosteus aculeatus consumed per crab in the 24, 48, 72, and 144 hour field multiple-prey experiments (Bare enclosures: solid line; eelgrass enclosures: dashed line; x±S.E., N=2). 25 Table 5. Summary of the repeated-measures ANOVA statistics comparing the num-ber of individuals consumed in the bare enclosures to that in the eelgrass enclosures for the field multiple-prey experiment Source of Variation DF MS Macoma nasuta Time Treatment Interaction 3 1 3 121.56 390.06 13.56 8.96 44 .89 1.56 0.054 0.0005 0.26 Crangon franciscorum Time 3 Treatment 1 Interaction 3 34.06 0.06 5.40 6.31 0.02 1.92 0.08 0.37 0.22 Gasterosteus aculeatus Time 3 Treatment 1 Interaction 3 6.23 5.06 6.23 0.00 2.07 2.55 1.00 0.21 0.14 Table 6. Summary of the repeated-measures ANOVA statistics comparing the num-ber of individual prey consumed in the bare with that of the eelgrass enclosures for the laboratory multiple-prey experiment Source of Variation DF MS Macoma nasuta Time Treatment Interaction 10 1 10 80.00 151.52 36.48 2.19 11.34 2.73 0.11 0.0005 0.010 Crangon franciscorum Time 10 Treatment 1 Interaction 10 36.93 180.02 3.05 12.11 36.15 0.61 0.0005 0.0005 0.69 Gasterosteus aculeatus Time 10 Treatment 1 Interaction 10 1.45 94 .56 0.36 4.03 189.12 0.72 0.021 0.0005 0.46 26 Virtually 100% of the shrimp were consumed in both types of enclosures within 24 hours (Fig. 6B) while fewer than 24% of the available fish were consumed after 144 hours (Fig. 6C). The number consumed of both shrimp and fish did not differ sig-nificantly between the two habitats (ANOVA, F=0.02 and F=2.07 respectively, p=0.05, Table 5). In the laboratory multiple-prey experiment, the consumption of all three prey species was significantly greater in the bare enclosure compared to that in the eelgrass enclosures (ANOVA, Macoma nasuta, F=11.34, p=0.0005; Crangon franciscorum, F=36.15, p=0.0005; Gasterosteus aculeatus, F=189.12, p=0.0005, Fig. 7, Table 6). The result for the clams is contrary to that obtained in the laboratory single-prey experiment. Crabs in the bare enclosures of both field and laboratory multiple-prey experiments consumed a significantly greater total wet biomass of prey than did crabs in the eelgrass enclosures (ANOVA, field, F=75.10, p=0.0005; laboratory, F=224.00, p=0.0005, Fig. 8, Table 7). Although the relationship of biomass consumed per crab over time appears to differ between the field and laboratory experiments, this is primarily due to experimen-tal differences between the field and the laboratory and therefore are not directly com-parable. In the field, four independent experiments varying in duration were combined to examine the relationship between biomass consumed per crab over time while in the laboratory, the relationship represents one, continuously monitored, experiment. 27 A. Macoma nasuta i i — — — — i 1 T — — — r O 24 48 72 86 120 144 Time (Hours) Figure 7. Number of individual Macoma nasuta, Crangon franciscorum, and Gasterosteus aculeatus consumed per crab over 144 hours in the bare (solid line) and eelgrass (dashed line) of the laboratory multiple-prey experiment (x±S E N=3) 28 A. Field Multiple-Prey Experiment 16 • J3 CO o o a. | ^ 0 -I . ' ' r c • O CO B. Laboratory Multiple-Prey Experiment CO CO E o m E CO L_ O Time (Hours) Figure 8. Biomass consumed per crab in bare (solid line) and eelgrass (dashed line) enclosures of a) field and b) laboratory multiple-prey experiments (x+S E • field, N=2; laboratory, N=3). ~ *' 29 Table 7. Summary of the repeated-measures A N O V A statistics comparing the gram biomass consumed in the bare enclosures with that of the eelgrass enclosures for the field and laboratory multiple-prey experiment S o u r c e o f V a r i a t i o n D F M S F p F i e l d M u l t i p l e - P r e y E x p e r i m e n t s Time 3 24.34 6.13 0.09 Treatment 1 59.33 75.10 0.0005 Interaction 3 5.03 5.03 0.032 L a b o r a t o r y M u l t i p l e - P r e y E x p e r i m e n t Time 10 75.67 19.71 0.0005 Treatment 1 224.00 29.47 0.0005 Interaction 10 3.84 0.51 0.54 30 DISCUSSION Reduced predation in seagrass meadows has been shown in a variety of studies to be an important influence on the composition of seagrass communities (Orth, 1977; Young and Young, 1977; Virnstein, 1978; Nelson, 1979a, 1981, Leber, 1985: Lewis, 1984; Summerson and Peterson, 1984: Virnstein et al, 1984; Main, 1987). Although the mechanism of influence differs depending on the species involved and on geographic location, these studies demonstrate that in providing a refuge from predation, seagrass meadows maintain a diverse faunal assemblage which in a less complex habitat would be eliminated, or at least reduced, by predation. Within seagrass communities, some groups appear to be more susceptible to preda-tion than others. This susceptibility may be governed by the prey's choice of microhabitat, making it more inaccessible to a predator, or by a more active ability of the prey to avoid predation. In summarizing the literature on predator-prey relationships in seagrass meadows, Orth et al (1984) concluded that susceptibility to predation in seagrass habitats was largely determined by the microhabitat which the prey species oc-cupied. In general, epifauna appeared to be more susceptible to predation than infauna and, among the epifauna, highly motile species tended to be less susceptible than less motile ones. As one example, in a Florida seagrass meadow, pinfish Lagodon rhom-biodes and shrimp Palaemonetes vulgaris were observed by Nelson (1970a,b; 1981) to prey more heavily on epifaunal amphipods than on free-living or infaunal amphipod species. The susceptibility to predation of the three prey species used in the present study in general appeared to be determined by the type of microhabitat occupied within the 31 seagrass meadow. In the field multiple-prey experiment, the epifaunal Crangon fran-ciscorum was much more susceptible to predation than were either the infaunal Macoma nasuta or neritic Gasterosteus aculeatus. In both bare and eelgrass habitats, shrimp were consumed quickly and in relatively high numbers, even when alternative prey were present (see Fig. 6). Thus, although the shrimp are potentially able to detect the predator by visual, olfactory and tactile means, they appear to be vulnerable to predation. Shrimp were unable to maintain more than a short period of time swimming in the water column. Between these active periods, they remained slightly buried in the substrate and never attached to the eelgrass shoots. While resting on the substrate, shrimp may be readily accessible to predators and this behaviour may increase their vulnerability to predation relative to the other two prey species. In contrast, the infaunal Macoma nasuta appeared to have the lowest susceptibility to predation by Cancer magister. This species escaped predation by burrowing into the sediment where it remained hidden from, and/or inaccessible to, the foraging crab. In the field enclosures, there was little or no predation on the clam until after consumption of alternative prey species had ceased. This result parallels the observation by Blundon and Kennedy (1982) that consumption of the clam Mya arenaria by the blue crab Cal-linectes sapidus decreased when an artificial rhizome mat was present. Similarly, Peter-son (1982) attributed the increased survival of two infaunal seagrass inhabiting bival-ves, Chione cancelata and Mercenaria mercenaria, to the combination of rhizomes and increased compaction of sediments associated with these root systems. When the field and laboratory results of my study are compared, a clear illustration is gained of how the rhizome mat, rather than the eelgrass shoots, reduces crab predation. In the laboratory experiments the loose nature of the rhizomes, even in the presence of artificial eelgrass shoots, provided little apparent obstruction to foraging activities of the crab. In com-parison, the natural rhizome mat in the field eelgrass enclosures, being more tightly in-terlocked, significantly reduced crab predation. 3 2 . In comparison to Crangon franciscorum, few Gasterosteus aculeatus were con-sumed in any of the field or laboratory experiments, and any loss to predation occurred early in the experiment. Laboratory results indicated that after a short period of ex-posure to the predator, stickleback were able to avoid completely any further predation. This avoidance of predation may have been achieved by maintaining a position above the reach of the crabs. The difference in susceptibility of the shrimp and fish to preda-tion may therefore simply be due to a more refined ability of stickleback to evade the predator. Thus, in addition to the influence of microhabitat on the susceptibility, it ap-pears that sticklebacks may, through a learned response, have an ability to modify their behaviour and thereby further reduce their susceptibility to predation. Main (1987) has illustrated how predator-avoidance behaviours play an important role in reducing preda-tion in seagrass meadows. He observed that the shrimp Tozeuma carolinense was able to alter its behaviourial patterns while in the presence of the predatory pinfish Lagodon rhomboides and thereby reduce the predation efficiency of this predator. Cancer magister was able to capture and consume a variety of prey in both eelgrass and bare habitats. While the presence of eelgrass generally reduced this ability, there were some obvious discrepancies between field and laboratory results. The field and laboratory single-prey experiments principally differed with respect to predation on Macoma nasuta. The field experiment showed that significantly more M. nasuta were consumed in the bare enclosures than in the eelgrass enclosures, while the laboratory experiment showed no significant difference. This discrepancy appeared to be primari-ly due to the difficulty in constructing an effective rhizome mat system in the laboratory. The rhizomes brought in from the field and assembled in the laboratory did not resemble the thick mat present in the field, and the loose nature of these rhizomes and sand sub-strate of the laboratory experiments seemed to allow the crabs easier access to the clams. For this reason, I believe the field results best illustrate the effect of habitat complexity 33 on crab predation on infaunal prey in the present study. The field and laboratory single-prey experiments also differed with respect to the extent of predation on Crangon franciscorum and Gasterosteus aculeatus. The laboratory experiment showed that significantly more shrimp and fish were consumed in the bare enclosures than in the eelgrass enclosures while the field experiment showed the opposite result for shrimp and showed that the consumption of fish was almost iden-tical in the bare and eelgrass enclosures. However, the field results were based solely on the end point of a 24 h experiment and allow no resolution of what occurred prior to this time. In the laboratory, where continual monitoring was possible, the rate at which these two prey species were consumed demonstrates how eelgrass slowed the rate of predation in comparison with that in the bare habitat. Over the first 24 h, however, the rate at which these prey species were consumed in the bare enclosures declined to zero, while prey continued to be consumed in the eelgrass enclosures (see Fig. 5b,c). If the experiment had continued beyond 72 hours, it is possible that the number of fish and shrimp consumed in the eelgrass enclosures may have equalled or exceeded the number which had been consumed in the bare enclosures. However, the first 12 h of both the shrimp and fish single-prey laboratory experiments probably more accurately represents the natural predation conditions for the crab, given that in an estuarine sys-tem the time available to the crab for foraging is limited by the movement of the tides. Cancer magister in seagrass meadows may trade-off the potential cost of reduced foraging efficiency for the benefits of abundant food, warm temperatures and a refuge from their own predators. Crabs and predatory decapods in general may rival higher predators such as fish in their ability to influence seagrass community structure. As in-termediate predators, crabs not only influence the abundance of their prey but are also influenced by the variety of predators which prey upon them. My study and others by Virnstein (1977), Nelson (1979a,b, 1980), Coen et al. (1981), Blundon and Kennedy 34 (1982), and Leber (1985) demonstrate a reduction in the foraging efficiency of decapods in seagrass meadows as compared with less complex habitats. However, none of these studies has examined how the foraging efficiency of decapods in seagrass meadows might differ while in the presence of their predator(s) or another competing predator. A related study on freshwater crayfish by Stein and Magnuson (1976) showed that this predatory decapod reduced its foraging activity while in the presence of its own predator (a fish). In an marine environment, a behaviourial response such as this would further limit the time for foraging activity beyond that already restricted by the movement of the tides. In addition, a competing predator will also reduce the foraging success of another predator by reducing the abundance of shared prey. In this regard, Nelson (1979a) found that removal of Lagodon rhomboides, a fish predator competing with the prawn Palaemonetes, from a seagrass enclosure resulted in an increase in the abundance of the predatory shrimp. Alternatively, a competing predator may indirectly enhance the predation success of another predator. Thus, Rahel and Stein (1988) reported that by forcing the johnny darter fish, Etheostoma nigrum, into crevices which were occupied by crayfish, Orconectes rusticus, the small-mouth bass, Micropterus dolomieui, indirect-ly increased the foraging success of the crayfish. The presence of a predator of a decapod, or of a competing predatory species in a seagrass meadow may therefore either further reduce the foraging efficiency of the decapod or indirectly enhance its foraging success, depending on the species involved. My study has examined the predation by Cancer magister in two habitats, eelgrass and bare. However, it is important to consider the heterogenous nature of seagrass meadows in that, within the meadows, predators have a continuum of habitats available, ranging from bare substrate to seagrass in high density. Although the extent to which crabs move from one type of habitat to another within the seagrass meadow is unknown, the crabs in my study were observed to occupy both eelgrass and bare habitats at Roberts 35 Bank. If crabs utilize a variety of habitats within a seagrass meadow, then from the results of my study it is possible to speculate that crabs, rather than foraging in a seagrass habitat where their individual predatory success is reduced, may move to adjacent bare substrate where they are able to consume more prey per unit time than in a seagrass habitat. After weighing the costs and benefits of predators inhabiting a seagrass meadow, Summerson and Peterson (1984) have proposed that epibenthic predators such as crabs may remain in seagrass habitats when the risk of predation upon them is the greatest, and then emerge to forage on adjacent bare substrate where their predation effort is more efficient. While foraging on bare substrate, Cancer magister may not only substantial-ly reduce the faunal abundance in this habitat, but may also directly force prey from these bare habitats into the more protective seagrass habitats. The ability of a prey, while in the presence of a predator, to switch from one habitat to another providing greater protection has been documented in a variety of freshwater and marine habitats by Stein (1977), Sih (1982), and Main (1987). Therefore the access of predators to both bare and seagrass habitats may result in intense predation in the more foraging-efficient bare habitat. This factor coupled with a shift of the more motile prey species to the less predator-vulnerable eelgrass habitat may explain how seagrass habitats maintain their rich and abundant faunal assemblages. 36 References Armstrong, D. A. and D. R. Gunderson. 1985. The role of estuaries in Dunge-ness crab early life history: a case study in Grays Harbor, Washington, p. 145-170, Alaska Sea Grant Report No. 85-3, Alaska University, Proceedings of the Sym-posium on Dungeness Gab Biology and Management, Lowell Wakefield Symposia Series 1985. Arnold, W. S. 1984. The effects of prey size, predator size and sediment com-position on the rate of predation of the blue crab Callinectes sapidus on the hard clam Mercenaria mercenaria. J. Exp. Mar. Biol. Ecol. 80: 207-220. Bell, J. D. and M. Westoby. 1986. Abundance of macrofauna in dense seagrass due to habitat preference, not predation. Oecologia 68: 205-209. Blundon, J. A. and V. S. Kennedy. 1982. Refuges for infaunal bivalves from blue crab, Callinectes sapidus (Rathbun), predation in Chesapeake Bay. J. Exp. Mar. Biol. Ecol. 65: 67-81. Botsford, L. W. and D. E. Wickham. 1978. Behaviour of an age-specific, den-sity-dependent model and the northern California Dungeness crab (Cancer magister) as an example. Can. J. Fish. Aquat. Sci. 43: 2345-352. Butler, T. H. 1954. Food of the commercial crab in the Queen Charlotte Islands region. Fish. Res. Bd Canada, Pac. Prog. Rept. 99: 3-5. Butler, T. H. 1956. The distribution and abundance of early post-larval stages of the British Columbia commercial crab. Fish. Res. Bd Canada, Pac. Prog. Rept. 107: 22-23. Butler, T. H. 1980. Shrimps of the pacific coast of Canada. Can. Bull. Fish. Aquat. Sci. 202: 280p. Coan, E. V. 1971. The northwest Tellinidae. Veliger 14 (Supplement): 63p. Coen, L. D., K. L. Heck, and L. G. Abele. 1981. Experiments on competition and predation among shrimps of seagrass meadows. Ecology 62: 1484-1493. Eckman, J. E. 1984. Hydrodynamic processes affecting benthic recruitment. Lim-nol. Oceanogr. 28: 241-257. Eckman, J. E. 1987. The role of hydrodynamics in recruitment, growth, and sur-vival of Argopecten irradians (L.) and Anomia simplex (D'Orbigny) within eelgrass meadows. J. Exp. Mar. Biol. Ecol. 106: 165-191. Gee, J. M., R. M. Warwick, J. T. Davey, and C. L. George. 1985. Field experi-ments on the role of epibenthic predators in determining benthic community struc-ture. Ecol. 68: 1856-1862. 37 Gotshall, D. W. 1977. Stomach contents of northern California Dungeness crabs, Cancer magister. Calif. Fish and Game 63: 43-51. Greening, H. S. and R. J. Livingston. 1982. Diel variation in the structure of seagrass-associated epibenthic macroinvertebrate communities. Mar. Ecol. Prog. Ser. 7: 147-156. Hall, M. O. and S. S. Bell. 1988. Response of small motile epifauna to com-plexity of epiphytic algae on seagrass blades. J. Mar. Res. 46: 613-630. Hart, J. L. 1973. Pacific fishes of Canada. Bull. Fish. Res. Bd. Canada 180: 740p. Hart, J. F. L. 1982. Crabs and their relatives of British Columbia. B. C. Provin-cial Museum Handbook 40: 266p. Heck, K. L. Jr., and T. A. Thoman. 1981. Experiments on predator-prey interac-tions in vegetated aquatic habitats. J. Exp. Mar. Biol. Ecol. 53: 125-134. Holland, A. F., N. K. Mountford, M. N. Hiegel, K. R. Kaumeyer, and J. A. Mihursky. 1980. Influence of predation on infaunal abundance in Upper Chesapeake Bay, U. S. A. Mar. Biol. 57: 221-235. Jamieson, G. S. 1986. Implications of fluctuations in recruitment in selected crab populations. Can. J. Fish. Aquat. Sci. 43: 2085-2098. Jensen, K. T. and J. N. Jensen. 1985. The importance of some epibenthic predators on the density of juvenile benthic macrofauna in the Danish Wadden Sea. J. Exp. Mar. Biol. Ecol. 89: 157-174. Leber, K. M. 1985. The influence of predatory decapods, refuge and microhabitat selection on seagrass communities. Ecol. 66: 1951-1964. Lewis, F. G. in. 1984. Distribution of macrobenthic crustaceans associated with Thallassia, Halodule and bare substrate. Mar. Ecol. Prog. Ser. 19: 101-113. Lewis, F. G. Ill, and A. W. Stoner. 1983. Distribution of macrofauna within seagrass beds: An explanation for patterns of abundance. Bull. Mar. Sci. 33: 296-304. Lough, G. 1976. Larval dynamics of the Dungeness crab, Cancer magister, off the central Oregon coast, 1970-1971. Fish. Bull. 74: 353-375. Main, K. L. 1987. Predator avoidance in seagrass meadows: prey behaviour, microhabitat selection and cryptic coloration. Ecol. 68: 170-180. Nelson, W. G. 1979a. Experimental studies of selective predation on amphipods: consequences for amphipod distribution and abundance. J. Exp. Mar. Biol. Ecol. 38: 225-245. Nelson, W. G. 1979b. An analysis of structural pattern in an eelgrass (Zostera marina) amphipod community. J. Exp. Mar. Biol. Ecol. 39: 231-264. 38 Nelson, W. G. 1981. Experimental studies of decapod and fish predation on seagrass macrobenthos. Mar. Ecol. Prog. Ser. 5: 141-149. Orth, R. J. 1977. The importance of sediment stability in seagrass communities, p. 281-300. In B. C. Coull [Ed.], Ecology of marine benthos. Univ. South Carolina Press, South Carolina, U.S.A. Orth, R. J., K. L. Heck Jr., and J. Van Montfrans. 1984. Faunal communities in seagrass beds: a review of the influence of plant structure and prey characteristics on predator-prey relationships. Estuaries 7: 339-350. Peterson, C. H. 1982. Clam predation by whelk (Busycon spp.): experimental tests of the importance of prey size, prey density and seagrass cover. Mar. Biol. 66: 159-170. Peterson, C. H. and R. Black. 1986. Abundance patterns of infaunal sea anemones and their potential benthic prey in and outside seagrass patches on a western Australian sand shelf. Bull. Mar. Sci. 38: 498-511. Peterson, C. H., H. C. Summerson, and P. B. Duncan. 1984. The influence of seagrass cover on population structure and individual growth rate of a suspension feeding bivalve, Mercenaria mercenaria. J. Mar. Res. 42: 123-138. Phillips, R. C. 1984. The ecology of eelgrass meadows in the Pacific Northwest: a community profile. U.S. Fish Wildl. Serv. FWS/OBS-84/24. 85p. Posey, M. H. 1988. Community changes associated with the spread of an intro-duced seagrass, Zostera japonica. Ecol. 69: 974-983. Rahel, F. J. and R. A. Stein. 1988. Complex predator-prey interactions and predator intimidation among crayfish, piscivorous fish and small benthic fish. Oecologia. 75: 94-98. Reise, K. 1978. Experiments on epibenthic predation in the Wadden Sea. Hel-golander wiss Meeresunter. 31: 55-101. Sih, A. 1982. Foraging strategies and the avoidance of predation in an aquatic in-sect Notonecta hoffinanni. Ecol. 63: 786-796. Smith, L. D. and B. C. Coull. 1987. Juvenile spot (Pisces) and grass shrimp preda-tion on meibenthos in muddy and sandy substrata. J. Exp. Mar. Biol. Ecol. 105: 123-136. Stein, R. A. 1977. Selective predation, optimal foraging and predator-prey inter-action between fish and crayfish. Ecol. 58: 1237-1253. Stein, R. A. and J. J. Magnuson. 1976. Behavioral response of crayfish to a fish predator. Ecol. 57: 751-761. Stevens, B. G. and D. A. Armstrong. 1984. Distribution, abundance, and growth of juvenile Dungeness crabs, Cancer magister, in Grays Harbor estuary, Washington. Fish. Bull. 82: 469-483. 39 Stevens, B. G., D. A. Armstrong, and R. Cusimano. 1982. Feeding habits of the Dungeness crab Cancer magister as determined by the index of relative importance. Mar. Biol. 72: 135-145. Stoner, A. W. 1980. The role of seagrass biomass in the organization of benthic macrofaunal assemblages. Bull. Mar. Sci. 30: 537-551. Stoner, A. W. 1982. The influence of benthic macrophytes on the foraging be-haviour of the pinfish Lagodon rhomboides (Linnaeus). J. Exp. Mar. Biol. Ecol. 58: 271-284 Summerson, H. C. and C. H. Peterson. 1984. Role of predation in organizing ben-thic communities of a temperate-zone seagrass bed. Mar. Ecol. Prog. Ser. 15: 63-77. Virnstein, R. W. 1978. The importance of predation by crabs and fishes on ben-thic infauna in Chesapeake Bay. Ecol. 58: 1199-1217. Virnstein, R. W., W. G. Nelson, F. G. Lewis, and R. K. Howard. 1984. Latitudinal patterns in seagrass epifauna: Do patterns exist, and can they be explained? Es-tuaries 7: 310-330. Young, D. K. and M. N. Young. 1977. Community structure of the macroben-thos associated with seagrass of Indian River estuary Florida, p 359-381. In: B. C. Coull [Ed.], Ecology of marine benthos. Univ. South Carolina Press, Colum-bia. Young, D. K., M. A. Buzas, and M. W. Young. 1976. Species densities of mac-robenthos associated with seagrass: a field experimental study of predation. J. Mar. Res. 34: 577-592. "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0097464"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Zoology"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "The effect of structural complexity in eelgrass meadows on the predatory activity of the Dungeness crab Cancer magister"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/27476"@en .