PREDATION ON PLANKTONIC MARINE INVERTEBRATE LARVAE
by
KEVIN BRETr JOHNSON
A DISSERTATION
Presented to the Department of Biology
and the Graduate School of the University of Oregon
in partial fulfillment of the requirements
for the degree of
Doctor of Philosophy
June 1998
11
"Predation on Planktonic Marine Invertebrate Larvae," a dissertation
prepared by Kevin B. Johnson in partial fulfillment of the requirements for
. .
the Doctor of Philosophy degree in the Department of Biology. This
dissertation has been approved and accepted by:
L5!lka (i8
Date
Committee in charge: Dr. Alan L. Shanks, Chair
Dr. William E. Bradshaw
Dr. Stephen S. Rumrill
Dr. Lynda P. Shapiro
Dr. Cathy Whitlock
Accepted by:
Vice Provost and Dean of the Graduate School
CURRICULUM VITAE
NAME OF AUTHOR: Kevin Brett Johnson
PLACE OF ~IRTH: Fullerton, California
DATE OF BIRTH: December 31, 1968
GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED:
University of Oregon
Brigham Young University
DEGREES AWARDED:
Doctor of Philosophy in Biology, 1998, University of Oregon
Bachelor of Science in Zoology, 1992, Brigham Young University
AREAS OF SPECIAL INTEREST:
Larval Ecology
Invertebrate Zoology
Marine Biology
Life-History Evolution
Natural History
PROFESSIONAL EXPERIENCE:
Research Assistant, Oregon Institute of Marine Biology and
Department of Biology, University of Oregon, Eugene, 1994-1998
Graduate Teaching Fellow, Oregon Institute of Marine Biology and
Department of Biology, University of Oregon, Eugene, 1992-1997
Teaching Assistant, Zoology Department, Brigham Young
University, Provo, 1991-1992
Missionary, Church of Jesus Christ of Latter-Day Saints, 1988-1990
v
AWARDS AND HONORS:
Western Society of Naturalists Paper Competition, Honorable
Mention, 1997
Department of Defense, Fellowship Program, Honorable Mention,
1995,
Brigham Young University, 'Y' scholar, 1987
Boy Scouts of America, Eagle Scout, 1986
International Thespian Society, Inductee, 1985
GRANTS:
American Museum of Natural History, 1995
VI
VB
PUBLICATIONS:
Bradshaw, W. B., and K. B. Johnson. 1995. Initiation of
Metamorphosis in the Pitcher-Plant Mosquito: Effects of Larval
Growth History. Ecology 76:2055-2065.
. .
Johnson, K. B. 1998a. The Nemertea. In press in A. L. Shanks,
editor. Identifying Planktonic Invertebrate Larvae of the Coos Bay
Estuary. Oregon Institute of Marine Biology, Charleston, Oregon,
USA.
Johnson, K. B. 1998b. The Phoronida. In press in A. L. Shanks,
editor. Identifying Planktonic Invertebrate Larvae of the Coos Bay
Estuary. Oregon Institute of Marine Biology, Charleston, Oregon,
USA.
Johnson, K. B. 1998c. The Sipuncula. In press in A. L. Shanks,
editor. Identifying Planktonic Invertebrate Larvae of the Coos Bay
Estuary. Oregon Institute of Marine Biology, Charleston, Oregon,
USA.
Johnson, K. B., and L. A. Brink. 1998. Predation on bivalve veligers
by polychaete larvae. Biological Bulletin 194: In press.
Johnson, K. B., and A. L. Shanks. 1996. In situ measurements of
predation upon larvae of benthic marine invertebrates. The
American Zoologist 36:72A (Meeting abstract).
Johnson, K. B., and A. L. Shanks. 1997. The importance of prey
densities and background plankton in studies of predation on
invertebrate larvae. Marine Ecology Progress Series 158:293-296.
Smith, D. L., and K. B. Johnson. 1996. A Guide to Marine Coastal
Plankton and Marine Invertebrate Larvae, Second Edition.
KendalllHunt, Dubuque, Iowa, USA.
VIll
ACKNOWLEDGEMENTS
I express sincere thanks to the following individuals for their
contributions to this dissertation. Alan Shanks provided adept guidance
. .
and was an excellent advisor. My Dissertation Advisory Committee greatly
improved my research and publications. A special thanks goes to Steve
Rumrill, who first pointed out to me the need for research on larval
mortality. Barbara Butler was invaluable in acquiring reference material.
I appreciate the support of the staff, students, and faculty at the Oregon
Institute of Marine Biology. Bruno Pernet assisted in the culture and
maintenance of scaleworm larvae. The staff and facilities at Friday Harbor
Laboratories were instrumental to my field experiments. Funding for this
research was provided by an American Museum of Natural History
Lerner-Gray Fund award to KBJ and NSF award #OCE-9521093 to ALS.
Thanks to Lee Braithwaite at Brigham Young University for inspiring me,
guiding me, and introducing me to marine plankton and invertebrate
larvae. Jim and Bonnie Thompson have supported my work and I am
thankful. I would like to thank my parents, Royle and Sue Johnson, for
encouraging me to pursue my dreams and teaching me how. Finally, I
want to express appreciation to my wife Colleen for unfailing support and to
my daughter Bethany for helping me keep things in perspective.
TABLE OF CONTENTS
Chapter Page
I. GENERAL INTRODUCTION.. .. .. 1
Selected References 13
I I. THE IMPORTANCE OF PREY DENSITIES AND
BACKGROUND PLANKTON IN STUDIES OF
PREDATION ON INVERTEBRATE LARVAE. . .. . . ID
Abstract ID
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Methods 22
Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Selected References . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Bridge. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
III. PREDATION ON BIVALVE VELIGERS
BY POLYCHAETE LARVAE. . . . . . . . . . . . . . . . . . . . . 36
Abstract 36
Introduction.. ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . '37
Materials and Methods 40
Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Summary.................................. 52
Selected References . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Bridge. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
IV. THE IMPORTANCE OF ENCOUNTER RADIUS
AND PREY SWIMMING SPEED IN PLANKTONIC
ENCOUNTER MODELS. . . . . . . . . . . . . . . . . . . . . . . . . 58
Abstract 58
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Methods , . . .. 64
Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
IX
TABLE OF CONTENTS, Continued
Selected References . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Bridge. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
V. PREDATIO!'f ON PLANKTONIC MARINE AND
ESTUARINE INVERTEBRATE LARVAE. .. . . . . . .. . 79
Abstract 79
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Methods. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . ro
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Selected References . . . . . . . . . . . . . . . . . . . . . . . . . . 145
VI. CONCLUDING SUMMARY . . . . . . . . . . . . . . . . . . . . . 154
BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
x
LIST OF TABLES
Table Page
Chapter III
1. Mean Number of Crassostrea gigas Veliger
Larvae in Individual Guts of Predatory
Larval Polychaetes .......................... 48
Chapter V
1. Summary of Observational Experiments............
2. Summary of Seeded Predators in
Manipulation Experiments ................... 100
3. Mean Recovery of Marked Larvae in
Observational Experiments ................... 105
4. Predation on Marked Bivalve Veliger Larvae
from Observational Experiments............... 106
5. Background Plankton in Observational Experiments.. 109
6. Manipulation Experiment: Predators on
Marked Pluteus (P) and Marked Veliger
(V) Larvae by Treatment......................... 113
7. Predator Encounter Radiuses ( R) and Predator and
Prey Swimming Speeds (v and u, Respectively).......
8. Estimates of Marked Pluteus Encounters with
Predators in Observational Experiments............ 123
9. Estimates of Marked Veliger Encounters with
Predators in Observational Experiments............ 125
10. Encounter Estimates of Specified Predators
with Wild Invertebrate Larvae in
Corral Assemblages Over 24 H.................... 132
Xl
Xll
LIST OF FIGURES
Figure Page
Chapter II
1. Pre'dator-Induced Mortality as a Function
of Prey Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2. For Three Predator-Prey Combinations, Percent Prey
Mortality at Densities Selected Based Upon
Observed Predation in Prey Density
Experiments .
Chapter III
1. Veliger Predation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Chapter IV
1. Body Radius of Suspension-Feeding Trochophore
Larva ofA. vittata, Synonymous with
Encounter Radius R .
2. The Number of Captured Prey in Arctonoe vittata
Trochophores Vs. Log Prey Density .
3. The Number of Prey in Arcton6e vittata
Trochophores Vs. Log Prey Density .
4. Encounter Prediction Curves for the GS
and CR Encounter Models 71
Chapter V
1. In Situ Corrals: Deployment and Dimensions. . . . . . . 94
2. Field of View from Sample Sorting with Myriad
Phytoplankton and Background Plankton,
Viewed Under White Light. . . . . . . . . . . . . . . . . . . 103
XlII
LIST OF FIGURES, Continued
Figure
Chapter V
Page
I~
I
3. The Effects of Prey Density and ;Background
Plankton on Larval Mortality. . . . . . . . . . . . . . . . . . . . . 112
4. Mean Number of Bivalve Veligers Consumed by the
Hydromedusa Proboscidactyla flavicirrata
in Laboratory Experiments. . . . . . . . . . . . . . . . . . . . . . 115
5. Laboratory Predation by Proboscidactyla flavicirrata
on Small, Large, and Mixed (Large and Small)
Bivalve Veligers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
6. Capture of a Bivalve Veliger by the Heterotrophic
Dinoflagellate Noctiluca scintillans 139
1CHAPTER I
GENERAL INTRODUCTION
. Many marine and estuarine invertebrates have complex life cycles
and produce planktonic larvae which reside in the water column for hours
to months (Levin and Bridges, 1995). Larvae may develop from free-
spawned ova or be released from adults or egg cases after a period of
brooding or encapsulation. A single adult invertebrate can produce vast
numbers of these planktonic propagules. For example, during a single
spawning season one female sand dollar (Dendraster excentricus) can
spawn 3.8 x 105 eggs (Morris et aI., 1980), one female dungeness crab
(Cancer magister) can release 2.5 x 106 larvae (Morris et aI., 1980), a female
oyster (Crassostrea gigas) can spawn 55.8 x 106 ova (Galtsoff, 1964), and the
sunflower star Pycnopodia helianthoides may release as many as 160 x 106
eggs (Chia and Walker,1991). These larvae then develop in the plankton
until competent for settlement and metamorphosis. The numbers of
competent larvae present in local plankton correlates with recruitment to
benthic communities (Co.pnell, 1985; Gaines et aI., 1985; Roughgarden et
aI., 1991). These studies of supply-side ecology have investigated the
important relationship between planktonic larval supply and benthic
community composition. Recruitment to benthic populations can be
2determined by the supply of larvae available in the plankton (Roughgarden
et al., 1984; Connell, 1985; Gaines et al., 1985; Roughgarden et al., 1991).
The number of new recruits to the benthic adult assemblages can be
high-the barnacle Semibalanus balanoides was observed to settle in
densities reaching 215 individuals cm-2 (Connell, 1985). When compared to
the area's estimated propagule production, however, these newly
metamorphosed juveniles are few. Some studies attempt to estimate
mortality rates by contrasting propagule production with benthic
recruitment (see Rumrill, 1990 for review). These studies have estimated
mortality rates to be from 0.03 day·l in the cone snail Conus quercinus
(Perron, 1986) to 0.80 day·l in the clam Mya arenaria (Ayers, 1956), but
cannot distinguish between larval and early juvenile mortality. Many
planktonic mortality studies suffer from the drawbacks and potential biases
of anecdotal information and indirect evidence (Strathmann, 1985), limiting
available reliable knowledge of the sources and importance of mortality.
High mortality rates are expected, however, because invertebrate
populations are generally stable over time, and mortality must occur
between spawning or release and recruitment. Possible sources of
planktonic mortality include fertilization failure, starvation, lethal
temperatures, the absenC'e of the proper settlement substratum, transport
away from suitable settlement sites, predation on embryos and larvae, and
pathogens and genetic abnormalities (Thorson, 1946, 1950, 1966; Rumrill,
1990). Only pathogens and genetic abnormalities have not been
investigated.
I
J
I~
t
I~,
I
1-
3
I
Fertilization failure is potentially responsible for 0% (Sewell and
Levitan, 1992) to 100% (Babcock et aI., 1992; Brazeau and Lasker, 1992) of
propagule loss. Many organisms, however, exhibit behaviors or other
adaptations to help overcome potential low fertilization. Some of these
include synchronous spawning, aggregated spawning, increased egg size,
and spawning in shallows or pools (reviewed in Levitan, 1995). Field
studies have shown that it is possible to have nearly 100% fertilization of ova
released into subtidal marine systems (Sewell and Levitan, 1992). Even
when fertilization is relatively low (such as 10% fertilization success
observed for the sea cucumber Holothuria coluber, Babcock et aI., 1992), a
million eggs produced by a single spawning female would produce 1.0 x 105
planktonic larvae.
Successfully fertilized eggs may develop into larvae which
subsequently starve in the plankton. Most invertebrate larvae are
planktotrophic (= plankton-feeding) and require external nutrients to
complete metamorphosis. The role of starvation as a source of mortality
has been investigated for many invertebrate larvae (see reviews by Olson
and Olson, 1989; Boidron-Metairon, 1995). An additional factor that may be
important in assessing the threat of starvation is the ability of many larvae
to uptake dissolved orgw:i1c matter (PavilIon, 1976; De Burgh and Burke,
1983; Lucas et aI., 1986; Jaeckle and Manahan, 1988, 1989). Substantial
fractions of nutrients needed for metabolism and development can be
obtained by DOM uptake (Manahan and Wright, 1991; Manahan, 1983) and
offset nutritional stress when particulate food is unavailable (Boidron-
4Metairon, 1995). Nutritional resources in the field, whether dissolved or
particulate, are usually sufficient to prevent starvation (Olson, 1985, Olson,
1987; Strathmann, 1987; Gallager, 1988; Boidron-Metairon, 1995).
Many larvae are sensitive or intolerant of extremes or fluctuations in
temperature (Pechenik, 1987). Sensitivity and intolerance may take the
.
form of changes in behavioral or physiological activity, changes in
developmental rates, or actual mortality (Pechenik, 1987). In general,
lower temperatures depress developmental rates (Bayne, 1965; Scheltema,
1967; Lima and Pechenik, 1985; Harms, 1984), but extreme increases in
temperature can also slow growth (Scheltema, 1967; Kingston, 1974;
Leighton, 1974). The effects of changes in temperature have not been widely
studied, but some evidence suggests that depressed developmental rates at
lower temperatures may not fully recover when larvae experience an
increase in temperature (Beaumont and Budd, 1982). Little evidence of
direct mortality from natural-temperature extremes or fluctuations is
available. Temperature's influence on larval mortality, whether by direct
or indirect means, is potentially important and continues to receive
attention from investigators.
Offshore transport can potentially displace entire populations of
planktonic larvae and remove them from the proximity of suitable coastal
settlement sites. Evidence of transport-dependent recruitment includes
pulses of barnacle settlement that are correlated with the migration of an
upwelling front onto the shore (Farrell et aI., 1991; Roughgarden et aI.,
1991) and the occurrence of shoreward-propagating internal waves (Shanks
5and Wright, 1987). When larvae are transported away from suitable
settlement sites, mortality results from finite planktonic life-spans or other
agents of mortality (temperature, starvation, or predation) which may
affect larvae to different extents as their duration in the plankton is
prolonged. Many larvae can delay metamorphosis for weeks or months in
the absence of a suitable place to settle. In one extreme laboratory study,
veligers of the snail Fusitriton oregonensis remained planktonic larvae for
4 years in the absence of the proper settlement cue (M. Strathmann, pers.
communication). Assuming that the potential time for larval persistence
in the plankton is finite, transport away from a proper site may force
metamorphosis and settlement at a site where the juvenile cannot survive
or the resulting adult cannot effectively reproduce (Jackson and
Strathmann, 1981). If agents other than transport itself are responsible for
mortality, then a prolonged planktonic period due to the unavailability of
sites will result in mortality by other means. This is an active area of
research and promises to reveal much about larval ecology, the importance
of larval supply, and the potential influence of physical oceanography on
the biology of marine invertebrates (Jackson and Strathmann, 1981).
Planktonic invertebrate larvae can be consumed either by benthic
suspension-feeders or plRnktonic predators. Planktonic embryos and
larvae may encounter benthic suspension-feeding predators shortly after
release, incidentally during their planktonic life, or as they attempt to settle
and test the benthos for a suitable substratum. Benthic predators may form
a "wall of mouths" (Emery, 1973) and can make the acquisition of a
6settlement site a hazardous undertaking. Organisms associated with coral
reefs may consume as much as 60% of passing zooplankton, which
included the larvae of crustaceans, polychaetes, cnidarians, molluscs, and
echinoderms (Glynn, 1973). Suspension-feeding barnacles also inhibited
recruitment of colonial ascidians and bryozoans in field experiments
conducted by Young and Gotelli (1988). The anthozoans Alcyonium
siderium and Metridium senile captured and consumed planktonic
invertebrate larvae (Sebens and Koehl, 1984). Not all suspension-feeders,
however, consume invertebrate larvae. For instance, Bingham and
Walters (1989) found that settling larvae escaped predation by suspension-
feeding ascidians and Rumrill (1987) calculated the risk of predation, by 2
species of benthic suspension-feeders consuming Asterina miniata
brachiolaria larvae, to be 1.2% per saltation event (i.e., settlement or re-
suspension). Additional evidence of low predation by ascidians includes
their lack of effect on larval recruitment in a study by Young (1989).
Mortality of settling larvae by benthic suspension-feeders is clearly variable,
but much more investigation is necessary to determine the overall risk of
predation presented by benthic suspension-feeders.
Planktonic predators of invertebrate larvae have been studied in the
'V
laboratory, in the field through correlation of high predator abundance and
larval decline, and by gut content analysis of field-caught predators.
Laboratory experiments have investigated the following factors and their
effect on larval predation rates: antipredator defenses (Pennington and
Chia, 1984; Morgan 1987, 1989), developmental stage and post-contact
7behavioral responses (Rumrill et al., 1985; Pennington et al., 1986), larval
size (Pennington and Chia, 1984; Rumrill et al., 1985; Rumrill, 1987),
container size (Toonen and Chia, 1993), prey density (Rumrill et al., 1985;
Pennington et al., 1986; Johnson and Shanks, 1997, Johnson and Brink,
1998), and background plankton presence (Johnson and Shanks, 1997;
Johnson and Brink, 1998). In a review on larval mortality, however,
Rumrill points out a caution with regard to laboratory experiments on
predation:
An important limitation is that the majority of laboratory experiments
have been conducted in small containers at prey densities that are 2 to
3 orders of magnitude greater than natural densities of larvae in the
plankton. Direct extrapolation of mortality rates from laboratory
studies is unwarranted because rates of predation in the laboratory
are strongly dependent upon the size of the experimental container.
(Rumrill, 1990, p. 173)
In order for laboratory experiments to provide information that is directly
applicable to estimates of natural mortality, much more information must
first be collected about specific natural predator-prey relationships with
confirmation that the containers employed do not create artifacts.
Predation on planktonic larvae can be studied in the field by
identifying pelagic predators whose abundance is inversely correlated with
that of larvae. One example of this is the predatory ctenophore Mnemiopsis
leidyi, whose abundance has been negatively correlated with larval
abundance and recruitment of crustaceans and fish (Nelson, 1925; Burrell
and Van Engel, 1976; Cowan et al., 1994). This method requires, however,
the fortuitous monitoring of key predators and the need to assume that
8larval decline in the plankton is due, in whole or part, to predation by these
predators.
Another method of monitoring predation on planktonic invertebrate
larvae is by gut content analysis of potential predators. Indeed, predators
have been identified based upon their gut contents. Examples of
invertebrate larval predators identified in this manner include the
ctenophore Mnemiopsis leidyi (Nelson, 1925; Burrell and Van Engel, 1976),
the hydromedusa Phialidium sp. (McCormick, 1969), decapod larvae
(Lebour, 1922), and salmon fry (Bailey et aI., 1975). Unfortunately, some of
these predators may have consumed larvae in cod-end plankton buckets
and predation may be an artifact of collection. For example, chaetognaths
are known to feed unnaturally or at increased rates on plankton in
collection reservoirs (Feigenbaum and Maris, 1984).
Because predation in the plankton may be determined by opportunity,
or encounters between predators and prey (see laboratory evidence of
density-dependent predation- Rumrill et aI., 1985; Pennington et aI., 1986;
Johnson and Shanks, 1997, Johnson and Brink, 1998), many predators may
feed unnaturally when concentrated with potential prey in plankton
samples. Even if it is assumed, however, that the presence of larvae in
predator guts is not an artifact of collection, predator gut analysis has not
been an effective method for evaluating the impact of predation on larval
populations. When gut content studies have identified predators, the focus
has often been on the composition of the predator's diet. Invertebrate larvae
are a minor part of predator diets. However, the relative importance of
jI
I
fj
I
I
9
predation has not yet been determined for an individual larva throughout
the duration of larval life. Since data on the concurrent density of
planktonic prey is rarely offered, little can be said about the importance of
particular predators in the ecology of larval invertebrates. For example,
Bailey et al. (1975) observed that salmon fry had preyed upon decapod
larvae. Only 9% of salmon fry guts sampled, however, contained decapod
larvae. Decapods only represented 1% of the diet by volume. Decapod
larvae are not likely to be an important component of young salmon diets
and nothing is known of the potential impact on decapod populations by
salmon predators. Some hydromedusae of Phialidium sp. consume
invertebrate larvae, but less than 10% of predators sampled contained
larvae and, in those, larvae comprised less than 3% of identified prey
(McCormick, 1969). As with decapod larvae and salmon fry, these larvae
are not likely to be an important component of Phialidium sp. diets and
nothing is known of the potential impact on larval populations by this
predator. Because this data focuses on predator diets rather than larval
risk, important questions still remain. How important is planktonic
predation over a larva's planktonic life? What is the daily risk of predation
for an individual larva from all potential predators?
It is possible to evaluate predation risk using gut contents in
combination with known digestion rates and field densities of predators and
prey. This has been done to evaluate predation on adult copepods by larval
fish (Purcell, 1990) and on copepods, fish eggs, and fish larvae by
coelenterates (Purcell et al., 1994; Chandy and Greene, 1995). These results
10
cannot be extrapolated to predation on invertebrate larvae because these
predators may preferentially consume copepods, larval fish, or eggs, and
digestion times vary with prey type and size (Purcell, 1982; Chandy and
Greene, 1995). According to Purcell (1982), this combination approach to
evaluating in situ predation requires accuracy in measurements of
digestion times for particular prey, identification of digested prey,
converting size to dry weight and carbon, and determining predator and
prey densities from plankton tows. We would add that predator digestion
times for particular food types can vary tremendously depending on the
total amount of food in the gut. For instance, trochophore larvae of the
scaleworm Arctonoe vittata will pass bivalve veligers within 3 to 4 hours
when several veligers have been consumed and more are available. A lone
veliger in the gut ofA. vittata, however, may remain in the gut for as long
as a day (K. Johnson, pers. obs.). In spite of the potentially inaccurate
assumptions, estimates using gut contents, digestion times, and densities
may more accurately estimate field mortality than estimates based upon
laboratory predation studies (Purcell, 1982). Laboratory studies of predation
are potentially fraught with behavioral artifacts (Reeve 1977, 1980), but it is
unknown whether indirect field studies or laboratory experiments provide
the best estimate of field ihortality.
This doctoral dissertation investigates planktonic predation on
invertebrate larvae both in the laboratory and the field. Laboratory
experiments examine the hypotheses that 1) changes in prey density can
influence the proportion of prey consumed and 2) natural background
I
1
J
t
I
11
plankton (i.e., the natural suite of diverse plankton in whole, unfiltered
seawater) reduces or eliminates predation. The bulk of laboratory
experiments are described in chapters II and III. In chapter II, three
species were examined in the laboratory as predators on echinoid and
cirriped embryos or larvae. In chapter III, five larval polychaete species
.
representing 4 families were investigated as predators on bivalve larvae. In
both studies a general pattern emerged: predation was dramatically
reduced when prey were presented at natural prey densities and with
background plankton.
Chapter IV investigates the importance of predator encounter radius
and prey swimming speed in a planktonic predator-prey encounter model.
Encounter estimates of a simple predator with its prey are compared to
actual observations of predation. The predators and prey selected to
examine the model are the trochophore larvae of the scaleworm Arctonoe
vittata and the veliger larvae of the oyster Crassostrea gigas.
Chapter V details field studies and related laboratory investigations
of predation on invertebrate larvae. Most field studies were designed to
simply observe predation, expose predator identities, and determine
predation rates under near-natural conditions. These observational field
studies test the hypothesIS that populations of Invertebrate larvae suffer
significant predation in near-natural plankton assemblages. To examine
factors affecting predation rates, additional field and laboratory studies test
the hypotheses that 1) proportion of predation on a larval population
changes with prey density and 2) natural background plankton reduces
12
predation rates. Field experiments used natural assemblages, including a
diverse suite of potential predators, enabling me to directly determine the
predation risk for experimental larval populations. Corrals were
inoculated with marked and enumerated invertebrate larvae at the start of
24 h experiments. By marking prey, we could know initial prey densities,
retrieve larvae after the experiment, determine the number of survivors,
and identify the natural predators. Observations of predation are direct and
can be related directly to the potential impact of predation on experimental
populations of invertebrate larvae. Corral assemblages also included wild
(i.e., randomly caught and unmarked) invertebrate larvae at natural
densities. We were able to examine predation risk for captured wild larvae
using predator gut content analyses, known wild prey densities, and a
planktonic predator-prey encounter model. Finally, corrals were also used
to manipulate prey density and "background plankton" presence,
examining their effect on predation rates.
Three of the ensuing chapters (II, III, and V) have co-authors. I am
the primary author of all chapters. The second author of chapters II and V
is Alan L. Shanks, my doctoral advisor. The second author of chapter III is
Laura A. Brink, a fellow graduate student at the Oregon Institute of
Marine Biology. In chapter V I shared equal responsibility for the
development of methods with my co-author. Research, data analysis, and
writing for chapter V were primarily my responsibility. In my other co-
authored chapters, I was the principal investigator in all aspects of the
study.
13
Selected References
Ayers, J. e. 1956. Population dynamics of the marine clam, Mya arenaria.
Limnology and Oceanography 1:26-34.
Babcock, R. C., e. N. Mundy, J. Keesing, and J. Oliver. 1992. Predictable
and unpredictable spawning events: in situ behavioural data from free-
spawning coral reef invertebrates. Invertebrate Reproduction and
Development 22:213-228.
Bailey, J. E., B. L. Wing, and e. R. Mattson. 1975. Zooplankton abundance
and feeding habits of fry of pink salmon, Oncorhynchus gorbuscha, and
chum salmon, Oncorhynchus keta, in traitors cove, Alaska, with
speculations on the carrying capacity of the area. Fishery Bulletin 73:846-
861.
Bayne, B. L. 1965. Growth and the delay of metamorphosis of the larvae of
Mytilus edulis (L.). Ophelia 2:1-47.
Beaumont, A. R., and M. D. Budd. 1982. Delayed growth of mussel (Mytilus
edulis ) and scallop (Pecten maximus ) veligers at low temperatures.
Marine Biology 71:97-100.
Bingham, B. L., and L. J. Walters. 1989. Solitary ascidians as predators of
invertebrate larvae: Evidence from gut analyses and plankton samples.
Journal of Experimental Marine Biology and Ecology 131:147-159.
Boidron-Metairon, I. F. 1995. Larval nutrition. Pages 223-248 in L.
McEdward, editor. Ecology of Marine Invertebrate Larvae. CRe Press,
Boca Raton, Florida, USA.
Brazeau, D. A., and H. R. Lasker. 1992. Reproductive success in the
Caribbean octocoral Briareum asbestinum. Marine Biology 114:157-163.
Burrell, V. G. J., and W. A. Van Engel. 1976. Predation by and distribution
of a ctenophore, Mnemiopsis leid)'i A. Agassiz, in the York River
estuary. Estuarine and Coastal Marine Science 4:235-242.
'\,.,.1
Chandy, S. T., and e. H. Greene. 1995. Estimating the predatory impact of
gelatinous zooplankton. Limnology and Oceanography 40:947-955.
Chia, F.-S., and e. W. Walker. 1991. Echinodermata: Asteroidea. Pages
301-353 in A. C. Giese, J. S. Pearse and V. B. Pearse, editors.
Reproduction of Marine Invertebrates V. VI Echinoderms and
Lophophorates. The Boxwood Press, Pacific Grove, California, USA.
14
Connell, J. H. 1985. The consequences of variation in initial settlement vs.
post-settlement mortality in rocky intertidal communities. The Journal
of Experimental Marine Biology and Ecology 93:11-45.
Cowan, J. H., Jr., E. D. Houde, and J.E. Purcell. 1994. Predator-prey
interactions in the estuarine plankton community of the Chesapeake
Bay, USA. 37th Conference of the International Association for Great
Lakes Research and Estuarine Research Federation: Program and
Abstracts, IAGLR, Buffalo, New York 166:166.
De Burgh, M. E., and R. D. Burke. 1983. Uptak~ of dissolved amino acids by
embryos and larvae of Dendraster excentricus (Eschscholtz)
(Echinodermata: Echinoidea). Canadian Journal of Zoology 61:349-354.
Emery, A. R. 1973. Comparative ecology and functional osteology of fourteen
species of damselfish (Pisces: Pomacentridae) at Alligator Reef, Florida
Keys. Bulletin of Marine Science 23:649-770.
Farrell, T. M., D. Bracher, and J. Roughgarden. 1991. Cross-shelf
transport causes recruitment to intertidal populations in central
California. Limnology and Oceanography 36:279-288.
Feigenbaum, D. L., and R. C. Maris. 1984. Feeding in the Chaetognatha.
Oceanography and Marine Biology Annual Review 22:343-392.
Gaines, S., S. Brown, and J. Roughgarden. 1985. Spatial variation in larval
concentrations as a cause of variation in settlement for the barnacle,
Balanus glandula. Oecologia 67:267-272.
Gallager, S. M. 1988. Visual observations of particle manipulation during
feeding in larvae of a bivalve mollusc. Bulletin of Marine Science 43:344-
365.
Galtsoff, P. S. 1964. The American oyster Crassostrea virginica Gmelin.
Fishery Bulletin 64:1-480.
Glynn, P. W. 1973. Ecology of a Caribbean coral reef. The Porites reef-flat
biotope: Part II. Plankton community with evidence for depletion.
Marine Biology 22:1-2}~
Harms, J. 1984. Influence of water temperature on larval development of
-Elminius modestus and Semibalanus balanoides (Crustacea,
Cirripedia). HelgoHinder wiss. Meeresunters 38:123-134.
Jackson, G. A., and R. R. Strathmann. 1981. Larval mortality from offshore
mixing as a link between precompetent and competent periods of
development. The American Naturalist 118:16-26.
15
Jaeckle, W. B., and D. T. Manahan. 1988. Feeding in lecithotrophic larvae
of Haliotus rufescens: uptake of dissolved amino acids from seawater.
American Zoologist 28:167A.
Jaeckle, W. B., and D. T. Manahan. 1989. Growth and energy balance
during the development of a lecithotrophic molluscan larva (Haliotus
rufescens). Biological Bulletin 177:237-246.
Johnson, K. B., and ~. L. Shanks. 1997. The importance of prey den,sities
and background plankton in studies of predation on invertebrate larvae.
Marine Ecology Progress Series 158:293-296.
Johnson, K. B., and L. A. Brink. 1998. Predation on bivalve veligers by
polychaete larvea. Biological Bulletin In press.
Kingston, P. 1974. Some observations on the effects of temperature and
salinity upon the growth of Cardium edule and Cardium glaucum larvae
in the laboratory. Journal of the Marine Biological Association of the
United Kingdom 54:309-317.
Lebour, M. V. 1922. The food of planktonic organisms. Journal of the
Marine Biological Association of the United Kingdom 12:644-677.
Leighton, D. L. 1974. The influence of temperature on larval and juvenile
growth in three species of southern California abalones. Fishery Bulletin
72:1137-1145.
Levin, L. A., and T. S. Bridges. 1995. Pattern and diversity in reproduction
and development. Pages 1-48 in L. McEdward, editor. Ecology of Marine
Invertebrate Larvae. CRC Press, Boca Raton, Florida, USA.
Levitan, D. R. 1995. The ecology of fertilization in free-spawning
invertebrates. Pages 123-256 in L. McEdward, editor. Ecology of Marine
Invertebrate Larvae. CRC Press, Boca Raton, Florida, USA.
Lima, G. M., and J. A. Pechenik. 1985. The influence of temperature on
growth rate and length of larval life of the gastropod, Crepidula plana
Say. Journal of Experimental Marine Biology and Ecology 90:55-71.
Lucas, A., L. Chebob-Chalabi, and D. A. Aranda. 1986. Transition from
endotrophy to exotrophy in the larvae of Mytilus edulis L. Oceano!. Acta.
9:97-103.
Manahan, D. T. 1983. The uptake and metabolism of dissolved amino-acids
by bivalve larvae. Biological Bulletin 164:236-250.
16
Manahan, D. T., and S. H. Wright. 1991. Uptake of dissolved organic matter
from seawater by marine invertebrates: the "Grover C. Stephens' Era".
American Zoologist 31:3A.
McCormick, J. M. 1969. Trophic relationships of hydromedusae in Yaquina
Bay, Oregon. Northwest Science 43:207-214.
Morgan, S. G. 1987. Morphological and behavioral antipredatory
adaptations of decapod zoeae. Oecologia 73:393-400.
. .
Morgan, S. G. 1989. Adaptive significance of spination in estuarine crab
zoeae. Ecology 70:464-482.
Morris, R. H., D. P. Abbott, and E. Haderlie. 1980. Intertidal Invertebrates
of California. Stanford University Press, Stanford, California, USA.
Nelson, T. C. 1925. On the occurrence and food habits of ctenophores in New
Jersey inland coastal waters. Biological Bulletin 48:92-111.
Olson, R. R. 1985. In situ culturing of larvae of the crown-of-thorns starfish
Acanthaster planci. Marine Ecology Progress Series 25:207-210.
Olson, R. R. 1987. In situ culturing as a test for the larval starvation
hypothesis for the crown-of-thorns starfish, Acanthaster planci.
Limnology and Oceanography 32:985-904.
Olson, R. R., and M. H. Olson. 1989. Food limitation of planktotrophic
marine invertebrate larvae: does it control recruitment success? Annual
Review of Ecology and Systematics 20:225-247.
PavilIon, J. F. 1976. Etude de l'absorption de quelques substances
organiques dissoutes marquees au 14C et au 3H par les oeufs et les larvae
de certains invertebres marins. Actualites de Biochimie Marine,
Colloque GAB1M - CNRS:211-231.
Pechenik, J. A. 1987. Environmental influences on larval survival and
development. Pages 551-595 in A. C. Giese, J. S. Pearse and V. B. Pearse,
editors. Reproduction of Marine Invertbrates IX General Aspects:
Seeking Unity in Diversity. Blackwell Scientific Publications and the
Boxwood Press, Palo Alto and Pacific Grove, California, USA.
Pennington, J. T., and F. S. Chiao 1984. Morphological and behavioral
defenses of trochophore larvae of Sabellaria cementarium (Polychaeta)
against four planktonic predators. Biological Bulletin 167:168-175.
17
Pennington, J. T., S. S. Rumrill, and F. S. Chiao 1986. Stage-specific
predation upon embryos and larvae of the pacific sand dollar, Dendraster
excentricus, by 11 species of common zooplanktonic predators. Bulletin of
Marine Science 39:234-240.
Perron, F. E. 1986. Life history consequences of differences in
developmental mode among gastropods in the genus Conus. Bulletin of
Marine Science 39:485-497.
Purcell, J. E. 1982. Feeding and .growth of the siphonophore Muggiaea
atlantica (Cunningham 1893). Journal of Experimental Marine Biology
and Ecology 62:39-54.
Purcell, J. E. 1990. Soft-bodied zooplankton predators and competitors of
larval herring (Clupea harengus pallasi) at herring spawning grounds
in British Columbia. Canadian Journal of Fisheries and Aquatic
Sciences 47:505-515.
Purcell, J. E., D. A. Nemazie, S. E. Dorsey, E. D. Houde, and J. C. Gamble.
1994. Predation mortality of bay anchovy Anchoa mitchlli eggs and larvae
due to scyphomedusae and ctenophores in Chesapeake Bay. Marine
Ecology Progress Series 114:47-58.
Reeve, M. R. 1977. The effect of laboratory conditions on the extrapolation of
experimental measurements to the ecology of marine zooplankton. 5. A
review. Proceedings of the Symposium on Warm Water Zooplankton
(NIO) pp. 528-537.
Reeve, M. R. 1980. Comparative experimental studies on the feeding of
chaetognaths and ctenophores. Journal of Plankton Research 2:381-393.
Roughgarden, J., S. Gaines, and Y. Iwasa. 1984. Dynamics and evolution
of marine populations with pelagic larval dispersal. Pages 111-128 in R.
M. May, editor. Exploitation of Marine Communities. Springer-Verlag,
New York, New York, USA.
Roughgarden, J., J. T. Pennington, D. Stoner, S. Alexander, and K. Miller.
1991. Collisions of upwelling fronts with the intertidal zone: the cause of
recruitment pulses inlSarnacle populations of central California. Acta
Oecologica 12:35-51.
Rumrill, S. S. 1987. Differential Predation upon embryos and larvae of
Pacific echinoderms. Ph.D. Thesis, University of Alberta, Edmonton,
Canada.
Rumrill, S. S. 1990. Natural mortality of marine invertebrate larvae.
Ophelia 32:163-198.
i
I
I
18
Rumrill, S. S., J. T. Pennington, and F. S. Chiao 1985. Differential
susceptibility of marine invertebrate larvae: Laboratory predation of sand
dollar, Dendraster excentricus (Eschscholtz), embryos and larvae by
zoeae of the red crab, Cancer productus Randall. Journal of
Experimental Marine Biology and Ecology 90:193-208.
Scheltema, R. S. 1967. The relationship of temperature to the larval
development of Nassarius obsoletus (Gastropoda). Biological Bulletin
132:253-265.
Sebens, K. P., and M. A. R. Koehl. 1984. Predation on zooplankton by the
benthic anthozoans Alcyonium siderium (Alcyonacea) and Metridium
senile (Actiniaria) in the New Enland subtidal. Marine Biology 81:255-
271.
Sewell, M. A., and D. R. Levitan. 1992. Fertilization success during a
natural spawning of the dendrochirote sea cucumber Cucumaria
miniata. Bulletin of Marine Science 51:161-166.
Shanks, A. L., and W. G. Wright. 1987. Internal-wave-mediated shoreward
transport of cyprids, megalopae, and gammarids and correlated
longhsore differences in the settling rate of intertidal barnacles. Journal
of Experimental Marine Biology and Ecology 114:1-13.
Strathmann, R. R. 1985. Feeding and nonfeeding larval development and
life-history evolution in marine invertebrates. Annual Review of Ecology
and Systematics 16:339-361.
Strathmann, R. R. 1987. Larval feeding. Pages 465-550 in A. C. Giese, J. S.
Pearse and V. B. Pearse, editors. Reproduction of Marine Invertbrates
IX General Aspects: Seeking Unity in Diversity. Blackwell Scientific
Publications and the Boxwood Press, Palo Alto and Pacific Grove,
California, USA.
Thorson, G. 1946. Reproduction and larval development of danish marine
bottom invertebrates. Meddr Komm Danm Fisk -og Havunders serie
Plankton 4:1-523.
Thorson, G. 1950. ReprodUctive and larval ecology of marine bottom
invertebrates. Biological Review 25:1-45.
Thorson, G. 1966. Some factors influencing the recruitment and
establishment of marine benthic communities. Netherlands Journal of
Sea Research 3:267-293.
19
Toonen, R. J., and F. S. Chiao 1993. Limitations of laboratory assessments
of coelenterate predation: container effects on the prey selection of the
Limnomedusa, Proboscidactyla flavicirrata (Brandt). Journal of
Experimental Marine Biology and Ecology 167:215-235.
Young, C. M. 1989. Larval depletion by ascidians has little effect on
settlement of epifauna. Marine Biology 102:481-489.
Young, C. M., and N. J. Gotelli. 1988. Larval predation by barnacles: effects
on patch colonization in a shallow subtidal community. Ecology 69:624-
634.
Zimmerman, S. T. 1972. Seasonal Succession of Zooplankton Populations in
Two Dissimilar Marine Embayments on the Oregon Coast. Ph.D. Thesis,
Oregon State University, Oregon, USA.
CHAPTER II
THE IMPORTANCE OF PREY DENSITIES AND BACKGROUND
PLANKTON IN STUDIES OF PREDATION ON
INVERTEBRATE LARVAE
In accordance with the regulations and approval of the University of
Oregon Graduate School, this chapter is a reproduction of previously
published and co-authored material: Marine Ecology Progress Series Vol.
158: 293-296, Kevin B. Johnson and Alan L. Shanks, co-authors.
Abstract
Laboratory experiments investigating predation by plankton on
meroplanktonic invertebrate larvae often use unnaturally high densities of
prey in filtered seawater. Offering prey under these conditions, however,
can alter predator behavior and capture success, potentially creating
artifactual predator-prey relationships and predation rates. We conducted
laboratory experiments investigating the effect of a range of larval
invertebrate densities on predation rates. For the four predator-prey
v
combinations examined, there was no predation at natural prey densities
in filtered seawater. We then conducted predator-prey experiments in the
presence and absence of naturally occurring ambient plankton
("background plankton") at densities where predation had been observed in
21
filtered seawater. In most experiments, background plankton dramatically
decreased or eliminated predation which had been observed with
unnaturally high prey densities in filtered seawater.
Introduction
. .
Laboratory experiments investigating predation upon
meroplanktonic invertebrate larvae are often conducted using unnaturally
high densities of meroplanktonic prey in filtered seawater. Unnaturally
high prey densities can alter predator behavior, capture success, and food
preference. These density effects have been observed in other predator-prey
systems (e.g., Holling 1959, Krebs et al. 1977). To the best of our knowledge,
however, this is the first study directly examining the influence of prey
densities on predation of invertebrate larvae by planktonic predators.
Using filtered seawater for laboratory predation experiments, like
using unnaturally high prey densities, may also induce unnatural
predation. Planktonic predators may be generalists, feeding upon all
potential prey, including the naturally occurring ambient plankton
("background plankton"). Background plankton, including protists and
phytoplankton, are far more abundant than relatively rare meroplanktonic
invertebrate larvae. Bybtcupying or satiating the predator, or obscuring
larvae from detection, background plankton may reduce larval predation.
Alternatively, predators may specialize in feeding on prey other than the
type being offered. In either case, predators consuming prey in filtered
seawater may not do so in the presence of background plankton.
22
We conducted predation experiments, observing predation rates, in
filtered seawater over a range of prey densities, including near-natural and
unnaturally high densities. Using prey densities where predation was
observed in filtered seawater, we then conducted predation experiments
with and without background plankton.
Methods
Three predators (the zoea of the mud shrimp Upogebia pugettensis,
the leptomedusa Obelia sp., and an unidentified leptomedusa) and three
prey types (blastulae and plutei of the purple sea urchin Strongylocentrotus
purpuratus, and barnacle nauplii) were used to create four predator-prey
combinations. Some zoeae and hydromedusae are known to be predatory
(e.g., Rumrill 1987), but no information is available on the natural prey of
our selected predatory species. S. purpuratus were spawned and
maintained using standard techniques (Strathmann 1987). Blastulae were
approximately 120 mm long and plutei were 4-arm stage and
approximately 200 mm in length. Barnacle nauplii (body length 200 to 250
mm) and all predators were collected at high tide from near the mouth of
Coos Bay, Oregon (43°21'10" N, 124°19'50" W) by slowly towing a plankton net
equipped with a large blind cod-end (after Reeve 1981). Experiments began
within 24 hours of predator collection and were conducted on a roller table
(Omori & Ikeda 1984, Larson & Shanks 1996), which rolled 3-liter
cylindrical tanks at 1 rpm and prevented plankton from settling. Though
enclosed, plankton do not suffer oxygen depletion during the experimental
time frame (Larson & Shanks 1996). The roller table was maintained at 12
°C in a constant temperature room with a 14:10 Light:Dark cycle for 24
hours. Observations of predators and prey in roller tanks revealed them to
stay suspended in the water column and exhibit apparently normal
behavior. At the end of each experiment, predators and remaining prey
were collected, fixed, and counts of surviving larvae made using a
compound microscope. "Mortality" is based upon the lack of retrieval of
whole, unconsumed larvae and the difference in mortality between
treatments with and without predators is attributed to predation.
When treatments are stated to be different, we refer to a = 0.05 with
the Games & Howell (G&H) mean significant difference method of a
posteriori pairwise comparison of means (Sokal & Rohlf 1995), performed
after a significant Kruskal-Wallis ANOVA (K-W ANOVA). The G&H
method of comparing means is appropriate for heterogeneous variances
and small sample sizes.
Prey -Density Experiments
Experiments investigating the effect of variation in prey density on
predation were conducted in 1 mm-filtered seawater with four different
predator-prey combinations: mud shrimp zoea preying upon plutei, mud
shrimp zoea preying upon blastulae, unidentified leptomedusa preying
upon barnacle nauplii, arid Obelia sp. medusa preying upon blastulae.
Predator density was 1 tank-t. Three replicate treatments (predators
present) and controls (predators absent) were run for each prey density.
24
FIGURE 1. Predator-induced mortality as a function of prey density. A.
Mud shrimp zoea preying upon purple urchin plutei. B. Mud shrimp zoea
preying upon purple urchin blastulae. C. Unidentified leptomedusa
preying upon barnacle nauplii. D. Obelia sp. medusa preying upon purple
urchin blastulae. Columns with zero mean and variance are indicated by a
"0". Error bars represent the 95% confidence interval. Predator treatments
that are significantly different from their predator-less controls at a.=.05 are
marked with a star.
35
30
25
20
A.
*
*
B.
11- mortality without a predatorD-mortality with predator
c.
15 r- <.
10
~~ 5
~ 0 I 00
~
~ 35
*
250
00 00
D.
o
100
30
25
20
15
10
5
01 00 00 00 J.O -U 100 OLLL
1 3 5 10 50 1 3
Prey Density (# larvae r1)
5 10 50 83
~
Prey densities (Figure 1) ranged from near-natural to unnaturally high
densities. Published observations of larval urchin field densities (and, by
extrapolation, conservative urchin blastula densities) range from 0.08 to
0.39 1-1 (Zimmerman 1972, Cameron & Rumrill 1982, Emlet 1986, Rumrill
1987, Rumrill et al. 1985) and the highest reported density is only 0.74 P
(Miller i995). Natural urchin densities are represented in our experiments
as a density of 11-1• By contrast, densities of echinopluteus larvae used in
past laboratory predation experiments has often ranged from 25 to 500 t 1
(e.g., Rumrill et al. 1985, Pennington et al. 1986). Natural densities for
barnacle nauplii may be as high as 15 t 1 (Zimmerman 1972). Natural
nauplius densities are represented in our experiments as densities of 1, 3,
5, and 101-1. Our high density of 501-1 exceeds published observations and is
intended to be unnaturally high. At the end of each experiment, predators
and remaining prey were collected and fixed. Counts of surviving larvae
were made using a compound microscope.
Background Plankton Experiments
Predation experiments with and without background plankton were
conducted with three of the same predator-prey combinations used in the
previous experiments. :Bxperiments were run at prey densities where
predation was observed in the above-described prey density experiments
(Figure 2). Experiments with the Obelia sp. medusa preying upon blastulae
and the unidentified leptomedusa preying upon barnacle nauplii consisted of 5
'Z7
FIGURE 2. For three predator-prey combinations, percent prey mortality at
densities selected based upon observed predation in prey density
experiments (see Figure 1): A. Obelia sp. preying upon blastulae B.
Unidentified leptomedusa preying upon barnacle nauplii C. Mud shrimp
zoea preying upon plutei. In A and B, the five columns for each prey
density are (left to right): 1. prey in filtered seawater (fsw) 2. prey and
predator in fsw 3. prey and backgroUnd plankton (bgp) 4. prey and predator
with bgp 5. prey and bgp fixed immediately (retrieval control). The four data
columns for each prey density in C represent treatments 1-4 above.
Columns with zero mean and variance are indicated by a "0". Error bars
represent the 95% confidence interval. Treatments that are significantly
different from their respective control at a=.05 are marked with a star.
1
!,
I 28
~ A.
~
r- f-
~
-
I-
-
-
*
- rI-
~
0 000 0 000
50
B.
*
35
~ 30
~~ 25
~ 20
~ 15
~
10
5
o L......,,;l;.L--l.,,;;-.J:&:l~--.J
5
• -preY,fsw
o -prey + predator,
fsw
~ -prey, bgp
~ - prey + predator,
bgp
.~;;,.... - retrieval control
';",,,.-
......... (A+B only)
35
30
.0
~ 25
~ 20
~~ 15
10
5
o
1
!
I
I
i
50
c.
* **
treatments, three replicates each, at each selected prey density. The five
treatments were prey alone in filtered seawater, prey with a predator in
filtered seawater, prey alone with background plankton, prey with a
predator and background plankton, and larvae and background plankton
fixed at the onset of the experiment (a control for retrieval artifacts in the
presence of background plankton). The protocol for the experiment with the
mud shrimp zoea preying upon plutei was the same as those described
above, but lacked the background plankton control. Background plankton
were obtained by collecting whole seawater (unfiltered seawater with a
natural composition and density of plankton) from near the mouth of Coos
Bay at high tide.
Results
Prey Density Experiments
For all predator-prey combinations the percent predation varied with
prey density. For the zoea preying upon plutei and blastulae, predation was
significant only at prey densities of 10 and 50 1"1 (Figure lA) and 50 P
(Figure IB), respectively. With the unidentified leptomedusa as a predator
on barnacle nauplii (Figure IC), significant predation was only observed at
a prey density of 50 }"1. Significant predation was observed at prey densities
...
of 50 and 83 }"1 with Obelia sp. as the predator on blastulae (Figure ID).
1I
I
Background Plankton Experiments
When Obelia sp. was a predator upon blastulae (Figure 2A), mean
mortalities of 31% and 10% were observed in filtered seawater at prey
densities of 5 and 50 P, respectively. When background plankton was
present, however, mortality was completely eliminated at both of these prey
densities. The primary components of background plankton in this
experiment included four diatom species and the dinoflagellate Noctiluca
scintillans. Background invertebrate larvae found in relatively low
numbers included polychaete metatrochophores (Spionidae) and copepod
nauplii. When background plankton and larvae were fixed immediately,
the exact number of added blastulae were retrieved in all replicates,
suggesting there were no wild blastulae in the background plankton
medium. Only one prey density, 50 1-1, was examined for the unknown
leptomedusa preying upon barnacle nauplii (Figure 2B). At this prey
density, the mean mortality of 27% in filtered seawater was completely
eliminated by the addition of background plankton. The primary
components of background plankton in this experiment included two
diatom species (different from species in the first background plankton
experiment) and a variety of moderately abundant dinoflagellates. Pine
pollen was also common in this background plankton. The number of
barnacle nauplii retrieved when background and larvae were fixed
immediately was exactly the number added in two of the replicates. In the
third replicate, 98% of added barnacle larvae were recovered. AB with
blastulae, this suggests that there were no wild barnacle larvae in the size
range of those used as prey. For the mud shrimp zoea preying on plutei
31
(Figure 2C) at a prey density of 10 P, the presence of background plankton
significantly reduced predation from an average of 16% to 1%. At a prey
density of 501-1, however, the average predation in filtered seawater was
14% vs. 17% in the presence of background plankton. Background plankton
consisted of relatively abundant loricated ciliates, dinoflagellates of the
genus Protoperidinium, and a wide variety of diatoms. This experiment
lacked the treatment where background plankton and larvae were fixed
immediately to control for artifacts. Retrieval of larvae with background
plankton in the absence of a predator, however, was exactly 100% at 10 t 1
and slightly less than 100% at 50 t 1• Once again, this suggests that wild
plutei were not added to the experiment by the use of background plankton.
In all but this last predator-prey combination, background plankton
reduced or eliminated predation.
Discussion
For all predator-prey combinations examined, predator-induced
mortality tended to increase with prey density. Predation at natural prey
densities was often nonexistent. The fact that predation tended to occur
only at unnaturally high densities may be due to altered predator behavior,
increased capture success at high densities, or may simply be the result of
more frequent encounters-with prey. Only in the latter case can predation
rates at unnaturally high densities be extrapolated to the lower natural
densities. Altered predator behaviors resulting from high densities of prey,
such as prey switching and selectivity, and increased capture success (i.e.,
practice makes perfect) may be artifactually induced when unnaturally
32
high prey densities are used. The mechanism underlying prey density's
effect on predation rates has not been identified for these predator-prey
combinations. Natural prey densities should be used to prevent behavioral
artifacts from misleading investigators about the existence or strength of
predator-prey relationships.
In all but one case, even when prey densities were unnaturally high,
background plankton reduced or eliminated predation which had been
observed in filtered seawater. Background plankton may serve as alternate
food, occupying or satiating generalist predators. Background plankton
may also obscure larvae from detection or hinder their capture. Whatever
the mechanism, background plankton reduced the likelihood of these
predators consuming meroplanktonic invertebrate larvae and embryos.
Background plankton, a pervasive component of natural planktonic
systems, should be present in laboratory investigations of planktonic
predation.
Much of the information on predators of marine invertebrate larvae
comes from laboratory experiments which have utilized unnaturally high
prey densities and excluded background plankton. These experiments have
contributed to the idea that predation in the plankton may be a major cause
of larval mortality (Rumrill 1990, Morgan 1995). In this study we included
natural prey densities and background plankton in an attempt to make our
laboratory experiments more natural. We found that, under more natural
conditions, predation was eliminated or greatly reduced. Perhaps previous
laboratory experiments have given us a false impression of predation rates
in the plankton.
33
Selected References
Cameron RA, Rumrill SS (1982) Larval abundance and recruitment of the
sand dollar Dendraster excentricus in Monterey Bay, California, USA.
Mar BioI 71:197-202
Emlet RB (1986) Larval production, dispersal, and growth in a fjord: a case
study on larvae of the sand dollar Dendraster excentricus. Mar Ecol
Prog ~er 31:245-254
Holling CS (1959) The components of predation as revealed by a study of
small mammal predation of the European pine sawfly. Canad Entomol
91:293-320
Krebs JR, Erichsen JT, Webber MI, Charnov EL (1977) Optimal prey-
selection by the great tit (Parus major). Anim Behav 25:30-38
Larson ET, Shanks AL (1996) Consumption of marine snow by two species
of juvenile mullet and its contribution to their growth. Mar Ecol Prog Ser
130:19-28
Miller B (1995) Larval abundance and early juvenile recruitment of
echinoids, asteroids, and holothuroids on the Oregon coast. (MS Thesis,
University of Oregon)
Morgan SG (1995) Life and death in the plankton: larval mortality and
adaptation. In: McEdward L (ed) Ecology of Marine Invertebrate Larvae.
CRC Press, New York
Omori M, Ikeda T (1984) Methods in Marine Zooplankton Ecology. John
Wiley & Sons
Pennington JT, Rumrill SS, Chia, FS (1986) Stage-specific predation upon
embryos and larvae of the pacific sand dollar, Dendraster excentricus,
by 11 species of common zooplanktonic predators. Bull Mar Sci 39(2):234-
240
Reeve MR (1981) Large cod-end reservoirs as an aid to the live collection of
delicate zooplankton. Limnol Oceanog 26(3):577-580
Rumrill SS (1987) Differential predation upon embryos and larvae of Pacific
echinoderms. (PhD Thesis, University of Alberta, Edmonton)
Rumrill SS (1990) Natural mortality of marine invertebrate larvae. Ophelia
32(1-2):163-198
34
Rumrill SS, Pennington JT, Chia, FS (1985). Differential susceptibility of
marine invertebrate larvae: laboratory predation of sand dollar,
Dendraster excentricus (Eschscholtz), embryos and larvae by zoeae of
the red crab, Cancer productus (Randall). J Exp Mar BioI Ecol 90:193-208
Sokal RR, RohlfFJ (1995) Biometry, third edition. WH Freeman and
Company, New York
Strathmann M (1987) Reproduction and Development of Marine
Invertebrates of the Northe:r:n Pacific Coast. University of Washington
Press, Seattle
Zimmerman ST (1972) Seasonal Succession of Zooplankton Populations in
Two Dissimilar Marine Embayments on the Oregon Coast. (PhD Thesis,
Oregon State University)
35
Bridge
Chapter II describes laboratory experiments which manipulate prey
densities and background plankton to study predation on barnacle nauplii
echinoid embryos and larvae. Predators examined in chapter II include an
a~omuran zoea and two hydromedusae. Chapt~r III describes laboratory
experiments which are similar in respect to hypotheses, parameters
manipulated, and general method to those presented in chapter II. In
chapter III, however, polychaete larvae were examined as predators of
bivalve larvae. A long history of anecdotal references in the literature to
predation on bivalve veligers by polychaete larvae adds depth and interest to
this study. The results of experiments in chapter III agree with those in
chapter II-predation is reduced or eliminated when prey are presented at
natural densities with background plankton present.
36
CHAPTER III
PREDATION ON BIVALVE VELIGERS BY POLYCHAETE LARVAE
In accordance with the regulations and approval of the University of
Oregon Graduate School, this chapter is a reproduction of previously
published and co-authored material: Biological Bulletin Vol. 194: in press,
Kevin B. Johnson and Laura A. Brink, co-authors.
Abstract
Polychaete larvae from several families are thought to be natural
predators upon planktonic bivalve larvae. However, little direct evidence of
interactions between these predators and prey is available. We conducted
predator-prey experiments on laboratory roller tables for five putative
predatory polychaete larvae, representing four families (metatroch-Iess
larvae of the Polynoidae and metatrochophore larvae of the Spionidae, the
Magelonidae, and the Phyllodocidae). D-hinge veliger larvae of the oyster
Crassostrea gigas were offered as prey. Predation was monitored over a
range of prey densities and in the presence and absence of background
yo
plankton. "Background plankton" are any naturally occurring plankton
assemblages found in whole, unfiltered seawater at ambient
concentrations. For all polychaete larvae examined, when natural C. gigas
densities and background plankton were used, no predation was observed.
1
I
!
Magelonids and phyllodocids did not consume any C. gigas larvae,
regardless of conditions. Polynoid and spionid trochophores consumed C.
gigas veligers at both the "natural" and unnaturally high prey densities in
filtered seawater. The addition of background plankton eliminated the
predation at all natural prey densities and significantly reduced the
predation observed at high prey densi·ties.
Introduction
Predation in the plankton is a source of mortality which may control
the presence and abundance of the planktonic larvae of benthic marine
invertebrates (Thorson, 1950). Observations of predation upon
meroplanktonic invertebrate larvae are recorded from as far back as the
1920s. For example, Lebour (1922) noted bivalve veliger larvae in the guts of
the larval polychaete Magelona papillicornis (Magelonidae). Other
biologists have also observed bivalve veligers within the guts of field-caught
Magelona sp. larvae (Thorson, 1946; Smidt, 1951; Ktihl, 1974; Wilson, 1982).
Lebour (1922), Smidt (1951), and Kiihl (1974) recorded only bivalve larvae as
prey for magelonids, but Thorson (1946) and Wilson (1982) observed that M.
papillicornis also consumed other planktonic organisms. In spite of these
many observations and the general impression that larval polychaetes of
the genus Magelona are specialist predators of bivalve veligers (e.g., Todd et
al., 1996), a natural predator-prey relationship between larval polychaetes
and bivalve larvae has yet to be definitively shown. There are problems also
with the anecdotal nature of some past observations on wild-caught
38
plankton: when planktonic predators and prey are concentrated in the cod-
end of a plankton net for several minutes or more, as is usually the case
when plankton samples are being collected, it is not possible to differentiate
natural predation from that occurring in the cod-end under very abnormal
conditions, which we refer to as "artifactual predation".
Predation upon bivalve veligers by polychaete trochophores
(metatroch-Iess trochophores and metatrochophores) has also been
observed for representatives of other polychaete families, including the
Polynoidae (Yokouchi, 1991), the Nephtyidae (Mileikovski, 1959; Yokouchi,
1991), the Phyllodocidae (Yokouchi, 1991), and the Spionidae (Daro and
Polk, 1973; K.B. Johnson, unpubl. data). These observations of predation
are remarkable in two ways. First, it is very seldom that a larva has been
observed to be the primary food consumed by a planktonic suspension-
feeding predator that consumes its prey one individual at a time. Unlike
cases in which predators (e.g;, some scyphozoans and clupeid fish)
indiscriminately feed on many planktonic prey, consistent observations of a
given prey item in the gut of such a "single-particle predator" may indicate
a strongly specific predator-prey relationship and provide insight into
predator behavior. Second, bivalve veligers consumed by polychaete larvae
are often surprisingly lal'ge relative to the predator's body diameter and
apparent mouth size (see Fig. 1).
Examining the mechanism underlying particle ingestion by
polychaete larvae, Phillips and Pernet (1996) fed larvae of the polychaetes
Serpula vermicularis (Serpulidae) and Arctonoe vittata (Polynoidae)
polystyrene beads and plankton at a range of sizes. S. vermicularis larvae
were apparently not equipped to handle food particles greater than 12 Jlm in
diameter (Phillips and Pernet, 1996). A. vittata larvae less than 100 Jlm in
diameter were observed to ingest large particles (polystyrene beads and
phytoplankton) up to 60 Jlm in diameter, a common size for small bivalve
larvae. The larvae ofA. vittata, a scaleworm, likely include relatively large
particles in their natural diet. Does this diet include larval bivalves?
Bivalve veligers have been observed in the guts of field-caught polynoid
larvae (Yokouchi, 1991). Like the larvae of Magelona sp., the larvae of
polynoids and several other polychaete families may be natural predators
upon bivalve veligers.
We examined the potential predator-prey relationship between
several larval polychaetes and bivalve veliger larvae. The relationship was
examined using a combination of field observations (plankton samples) and
laboratory experiments. In plankton samples, trochophores representing
several families were observed with bivalve veligers in their guts. More
important for this study, however, field samples helped determine densities
used in laboratory experiments. Densities of predators and prey reflected
field densities from samples where predation was observed. Laboratory
experiments used five types of larval polychaetes as predators: A. vittata
(metatroch-Iess trochophore, Polynoidae), Magelona sp. (metatrochophore,
Magelonidae), and unidentified species from the families Polynoidae
(metatroch-Iess trochophore), Spionidae (metatrochophore) and
Phyllodocidae (metatrochophore). D-hinge veliger larvae of the oyster
I
I
1
1
-"I
I
I
40
Crassostrea gigas were offered as prey. Experiments were conducted at
two prey densities and in the presence or absence of background plankton.
The presence of background plankton [by which we mean naturally
occurring phyto- and zooplankton ever-present in the field but often
excluded in laboratory experiments] is potentially important because it may
.
act as a substitute food for predators or obscure prey from detection
(Johnson and Shanks, 1997).
Materials and Methods
Field Observations
During August 1994, plankton samples were collected from within 10
km of the shore of Duck, North Carolina. Using a 100-Jlm-mesh plankton
net and an on-board electric centrifugal pump, samples were collected for 3
minutes at 227.11 minute-I, for a final sample volume of approximately 680
liters. Between 3 and 5 sampling depths were chosen at each station,
depending upon the station depth. Mter pumping was complete, samples
were rinsed from the cod-ends and preserved with 10% CaCOa-buffered
formalin for later sorting. Plankton samples were sorted under a
dissection microscope with polarized light to aid in locating bivalves. For a
-..-
more detailed description of collection and sorting methods, see Brink
(1997).
Bivalve veligers were tallied when observed in the guts of predatory
polychaete larvae. The total density of bivalve larvae and polychaete larvae
41
was determined for each sample in which bivalve predation was observed.
These densities were considered when deciding upon predator and prey
densities to be used in the laboratory experiments described below.
Culture of Predators and Prey
Adult specimens of the scaleworm Arctonoe vittata, commensal with
the keyhole limpet Diodora aspera, were collected with their host from the
west shore rocky intertidal of San Juan Island, Washington. Individuals of
A. vittata were spawned and larvae were cultured using the methods
described by Phillips and Pernet (1996) with the addition of Coscinodiscus
radiatus (CCMP 310) as a food source. Fertilized eggs were cultured in 600-
ml beakers at densities of -500 P. Larvae approximately 21 days old were
used as predators in experiments.
All other larval polychaetes used as predators were collected at high
tide near the mouth of Coos Bay, Oregon, by slowly towing a 150-J.1m-mesh
plankton net equipped with a large, blind cod-end (Reeve, 1981). Pipettes (3-
mm-bore ) were used to immediately remove predators from the plankton
sample and isolate them in 250 ml of filtered seawater. Experiments began
within 6 hours of predator collection.
D-hinge veligers oflhe oyster Crassostrea gigas, 5 to 10 days old
(greatest linear dimension 70-90 J.1m), were used as prey in all laboratory
experiments. The oyster larvae were obtained from Whiskey Creek Oyster
Farms, Tillamook, Oregon, and maintained in I-gallon jars on a diet of
Isochrysis galbana and Rhodomonas sp.
I42
Roller Table Experiments
One laboratory experiment, with four treatments, was conducted for
each of the five species of larval polychaete (Table n. Two densities of prey
were used. The first prey density (treatments A and B) was designed to
approximate natural field concentrations ang was set at 33 bivalve larvae t 1
on the basis of the highest value we found in the literature (Carriker, 1951).
The second prey density (treatments C and D) was chosen to represent an
unnaturally high concentration (1000 1-1) and thus increase the likelihood
that the prey would be encountered and ingested by predators. Each prey
density was presented to predators in either filtered seawater (treatments A
and C) or with background plankton (treatments B and D). Background
plankton was collected by filling buckets with whole, unfiltered seawater at
the high tide immediately preceding the start of an experiment. To fill
background treatment tanks, the seawater in buckets was stirred gently,
suspending settled plankton, and then poured into tanks.
For each experiment, all treatments and replicates were conducted
simultaneously. Cylindrical 3-liter tanks (19 cm dia. x 10.5 cm ht.) were
placed on a roller table (Omori and Ikeda, 1984; Larson and Shanks, 1996)
maintained at 12 ooC in a constant temperature room with a 14:10
light:dark cycle. The slow (1 rpm) rotation of the tanks kept the plankton
from settling, and the experiments were of short duration (24 h) to prevent
oxygen depletion (Larson and Shanks, 1996). At the close of the
experiments, the water in the roller table tanks was filtered through a
43
partially submerged 20-~m-meshNitex filter, and each tank was rinsed
twice to ensure that all polychaete larvae were retrieved. Within 2.5
minutes of filtration, polychaetes were located and isolated in filtered
seawater. Consumed bivalve larvae, visible through the polychaete larva's
transparent body, were then counted.
The experiment using Arctonoe vittata larvae as predators was
conducted at Friday Harbor Laboratories (Friday Harbor, Washington). A
predator density of 2 r 1 (6 tank-I) was chosen based upon the upper range of
polychaete trochophore densities from our field samples in which predation
upon bivalve larvae had been observed. Each tank was replicated three
times. Thus, a total of 18 polychaete larvae were used as predators for each
treatment.
All other experiments were conducted at the Oregon Institute of
Marine Biology (Coos Bay, Oregon). The four species of larval polychaetes
used as predators were Magelona sp. (metatrochophores) and three
unidentified species representing the families Polynoidae (metatroch-Iess
trochophores), Spionidae (metatrochophores), and Phyllodocidae
(metatrochophores). The unidentified genera will be referred to as polynoid
A, spionid A, and phyllodocid A, respectively. All predator densities in
"'~
Coos Bay experiments were 1 r 1 (3 tank-I) and, for each experiment, tanks
were replicated four times.
44
Results
Field Observations
Of 150 samples, 18 had at least one polychaete larva that had preyed
upon a bivalve veliger. A total of 30 bivalves were observed in the guts of 25
polychaete larvae (20 trochophores and 5 metatrochophores). The number of
bivalves consumed by each of the 20 metatroch-Iess trochophores was
variable: 1 trochophore larva had 3 bivalves, 2 trochophore larvae had 2
bivalves each, and 17 trochophore larvae had 1 bivalve each. Trochophores
were typically large (mean body length =237 J..lm, sd =35 J..lm) and robust in
form (for examples of body shape, see illustrations of polynoids,
phyllodocids, or nephtyids in Bhaud and Cazaux, 1987). Detailed
identification of these metatroch-Iess trochophores was often not possible,
but the following families may have been represented: Phyllodocidae,
Hesionidae, early Nephtyidae; Polynoidae, and Chrysopetalidae. Of those
metatrochophores which had bivalves, 3 were Magelona sp. with 1 bivalve
each. The last 2 metatrochophores were likely either phyllodocids or
hesionids; one (380 J..lm in length) had 2 bivalves in its gut, while the other
(368 J..lm in length) had 1 bivalve. In addition, a single metatroch-Iess
polychaete larva was observed with a gastropod veliger in its gut.
For the 18 samples in which bivalves were observed in polychaete
larva guts, densities ranged from 42 to 1193 polychaete larvae sample-1 (x =
277.2, sd = 324.3). The range of larval bivalve densities in these same
45
-1
samples was from 419 to 1949 larvae sample (x =1217.6, sd =494.2).
Therefore, at least 42 trochophores and 419 bivalve larvae were concentrated
together in the cod-end bucket (approximately 200 ml of seawater) when a
sample was complete.
Roller Table Experiments
Table 1 summarizes the results of the roller table experiments. For
the larvae of Magelona sp. and phyllodocid A, predation on bivalve veligers
was not observed in the laboratory under any conditions. The larvae of
Arctonoe vittata, polynoid A, and spionid A, however, did consume
Crassostrea gigas veligers (Fig. 1). These three polychaetes exhibited low
levels of predation when veliger larvae were presented at near-natural
densities and in filtered seawater (Table 1, Treatment A). When
background plankton was used with this same n,ear-natural prey density,
predation was always absent (Table 1, Treatment B). Predation was most
frequent when densities of C. gigas were high in filtered seawater (Table 1,
Treatment C). Notably, the polynoid larvae, A. vittata and polynoid A,
consumed the greatest numbers of veligers in Treatment C. The most
extreme was polynoid A, averaging 6.17 bivalve veligers gut-1 with two of the
""~...
individuals consuming 8 veligers each. Presenting prey at high densities
in the presence of background plankton (Table 1, Treatment D) reduced, but
did not eliminate, the predation observed at the same densities in
Treatment C.
46
FIGURE 1. Veliger predation. (A) D-hinge veliger of the oyster Crassostrea
gigas. (B) Trochophore larva of the polynoid Arctonoe vittata with a veliger
of the oyster C. gigas in its gut. (C) Metatrochophore larva of spionid A with
a C. gigas veliger in its gut. (D) Trochophore larva ofpolynoid A. with two
C. gigas veligers in its gut. A, C, and D are viewed with cross-polarized
light. Scale bar = 100 Jlm.
47
48
TABLE 1. Mean number of Crassostrea gigas veliger larvae in individual
guts of predatory larval polychaetes according to treatment (prey density
and the presence or absence of background plankton) ± the 95% CI.
Treatment
Near natural prey density High prey density
(33 prey 1-1) (l000 prey 1-1)
Larval Filtered Background Filtered Background
polychaete seawater plankton seawater plankton
(length) A B C D
Magelona sp. 0 0 0 0
(2-3 mm)
Phyllodocid A 0 0 0 0
(300-360 J.Uh)
A. vittata 1.05 ± 0.37 0 4.17±0.64 0.72 ± 0.38
(260-290 JlII1)
Polynoid A 0.83 ±0.41 0 6.17 ± 0.79 1.33±0.44
(280-310 JlII1)
Spionid A 0.08±0.16 0 1.33 ± 0.37 0.50 ± 0.38
(400-500 JlII1)
Polynoid trochophores, which consumed numerous veligers in
Treatment C, voided their gut contents through a large posterior rupture.
This rupture quickly heals and the unburdened trochophore suffers no
obvious permanent damage. Veliger valves sometimes remain attached at
the hinge after passage through the gut. Intact veligers which passed
through the guts of larval polychaetes were isolated in filtered seawater, but
no consumed veligers revived. Thus, while trochophore digestion can be
incomplete, predation does appear to result in mortality for bivalve larvae.
I
I
I
j
.~
49
Discussion
None of the larval polychaete species we tested consumed any bivalve
larvae when laboratory conditions were the closest to natural (i.e., near-
natural prey density with background plankton present; Table 1, Treatment
B). We d~d observe predation in the treatments which u~ed unnatural prey
density or filtered seawater. One explanation for the lack of predation in
Treatment B could be that larval polychaetes are not natural predators of
bivalve veliger larvae. In that case, previously published observations of
bivalve veligers in the guts of larval polychaetes might be an artifact of the
concentration of predators and prey in cod-end buckets during plankton
tows. Such artificial conditions can alter the behavior of predators and prey
and increase the probability of encounters between them, resulting in
unnatural ingestion. Cod-end predation is well documented for other
planktonic predators, such as chaetognaths (Feigenbaum and Maris, 1984),
and may mislead observers about predator-prey relationships.
Low encounter rates might also explain the absence of predation
under the most natural laboratory conditions used in this study. Predators
and prey may simply not encounter one another during the experiment.
Natural prey densities, which tend to be relatively low, and the presence of
background plankton can both decrease the number of encounters between
predators and prey (Johnson and Shanks, 1997). For example, lack of
encounters may explain the low predation by Arctonoe vittata on
Crassostrea gigas under the most natural conditions (Table 1, Treatment
B). This explanation is supported by comparisons between observed
predation by A. vittata and encounter model estimates (K.B. Johnson,
unpubl. data); the estimates produced by two models (Gerritsen and
Strickler, 1977, and a simple clearance rate model) were statistically
indistinguishable from the minimum known encounters of A. vittata with
C. gigas (i.e., observed predation events). This bolsters the argument that
larval polychaetes naturally prey upon bivalve veligers during relatively
infrequent encounters. Indeed, the many published observations of
predation (e.g., Thorson, 1946; Smidt, 1951; Kiihl, 1974; Wilson, 1982) may
reflect relatively rare field encounters rather than artifactual cod-end
predation. Predator-prey encounters in these previously published studies
can, however, be difficult to estimate. Field densities, swimming speeds,
and encounter radiuses, essential components of encounter rate models,
are often unknown. Finally, the hypothesis that these polychaetes may,
upon infrequent encounters, be natural predators of bivalve larvae is also
supported by an observation of a spionid larva with one C. gigas veliger in
its gut (K.B. Johnson, unpubl. data). This metatrochophore larva was fixed
only seconds after being collected in a 120-liter sample of seawater. No
plankton net was towed; the water was collected in a plastic bag, then
immediately concentrate'd and fixed. This method allowed little time for
artifactual predation.
The true frequency of encounters between predators and prey in the
field may, however, be far greater than estimated by models or from
laboratory experiments if natural densities are greater than those recorded
51
by investigators. The effect of plankton patchiness on sampling accuracy
has received some attention (Hamner and Carleton, 1979; Omori and
Hamner, 1982) and could cause underestimation of field densities.
Plankton can be highly concentrated in a localized area-for example,
through behavior-related aggregation (e.g., Alldredge and Hamner, 1980;
Ueda et aI., 1983) or the accumulation of plankton in a front (Stommel, 1949;
Bray, 1953; George and Edwards, 1973). A net, towed through such a patch
and then towed through a sparsely populated region, would collect a
sample with an apparent density lower than the actual density within the
front or aggregation. Furthermore, bivalve veligers are known to associate
with marine snow (Green and Dagg, 1997; Shanks and Walters, 1997),
creating localized high larval densities. Larval polychaetes can also be
strongly associated with marine snow (Shanks and del Carmen, 1997) and,
as a result, may encounter potential prey items such as bivalve veligers
more frequently. Published observations of predation upon bivalve veligers
by larval polychaetes may thus reflect natural predation in concentrated
patches of predators and prey.
In spite of the fact that we never observed predation on bivalve
veligers by Magelona larvae in laboratory experiments, published
observations of this predator-prey relationship are numerous and should
not be summarily dismissed. Wilson (1982) mentions that three species of
Magelona are known to be carnivorous in later stages and includes
descriptions of late stage metatrochophore larvae > 4 mm in length. The
Magelona metatrochophore larvae used in our experiments were 2-3 mm
52
long. At a later stage, with larger palps and mouths, these larvae may be
more effective at capturing bivalve larvae. It should be noted, however, that
a larva of Magelona papillicornis, lacking long palps and only 1 mm in
length, is depicted by Todd et al. (1996) with a bivalve veliger in its gut.
Experiments analogous to ours should be conducted with later stage
Magelona larvae to clarify the relationship of this predator with potential
bivalve prey.
Summary
Certain larval polychaetes may be significant natural predators upon
bivalve veligers. This investigation, however, provides laboratory evidence
that natural predation on bivalve larvae by polychaete larvae is absent or
uncommon, possibly because the predators and prey have few encounters
in the field (assuming that published larval bivalve densities accurately
reflect natural densities).
Published reports of bivalve veligers in the guts of larval polychaetes
suggest a natural predator-prey relationship and are seemingly
incongruous with our results. One possible explanation is that polychaete
larvae consumed the veligers while in the cod-end of a plankton net,
making the predation an"artifact of the collection method.
When polychaete larvae consumed bivalve veligers in our laboratory
experiments, the use of near-natural prey densities with natural
background plankton completely eliminated predation. This lack of
predation may be due to a reduction in the number of encounters with prey
53
(published data indicates that natural densities of bivalve larvae are
relatively low) or to the role of background plankton as a substitute food for
predators or a screen to obscure prey from detection. In short, our results
suggest that a natural predator-prey relationship between polychaete larvae
and bivalve veligers may not exist. If a relationship does exist, then the
frequency of interaction and its ecological importance may be less than
expected based upon published observations.
Selected References
Alldredge, A. L., and W. M. Hamner. 1980. Recurring aggregation of
zooplankton by a tidal current. Estuar. Coast. Mar. Sci. 10: 31-37.
Bhaud, M., and C. Cazaux. 1987. Description and identification of
polychaete larvae; their implications in current biological problems.
Oceanis 13: 596-753.
Bray, B. M. 1953. Sea-water discoloration by living organisms. N.Z.J. Sci.
Technol. B34: 393-407.
Brink, L.A. 1997. Cross-shelf transport ofplanktonic larvae of inner shelf
benthic invertebrates. M.S. Thesis, University of Oregon, Eugene.
Carriker, M. R. 1951. Ecological observations on the distribution of oyster
larvae in New Jersey estuaries. Ecol. Monogr. 21: 19-38.
Daro, M. H., and P. Polk. 1973. The autecology ofPolydora ciliata along the
Belgian coast. Neth. J. Sea Res. 6: 130-140.
Feigenbaum, D. L., and R. C. Maris. 1984. Feeding in the Chaetognatha.
Oceanogr. Mar. Biol.Ann. Rev. 22: 343-392.
George, D. G., and R. W. Edwards. 1973. Daphnia distribution within
langmuir circulations. Limnol. Oceanogr. 18: 798-800.
Gerritsen, J., and J. R. Strickler. 1977. Encounter probabilities and
community structure in zooplankton: a mathematical model. J. Fish.
Res. Board Can. 34: 73-82.
54
Green, E. P., and M. J. Dagg. 1997. Mesozooplankton associations with
medium to large marine snow aggregates in the northern Gulf of
Mexico. J. Plank. Res. 19: 435-447.
Hamner, W. H., and J. H. Carleton. 1979. Copepod swarms: attributes and
role in coral reef ecosystems. Limnol. Oceanogr. 24: 1-14.
Johnson, K. B., and A. L. Shanks. 1997. The importance of prey densities
and background plankton in studies of predation on invertebrate larvae.
Mar. Ecol. Prog. Ser. 158: 293-296.
Kiihl, H. 1974. Uber vorkemmen und nahrung der larven von Magelona
papillicornis O.F. Muller (Polychaeta Sedentaria) im Mundungsgebiet
von elbe, weser und ems. Ber. dt. wiss. Kommn. Meeresforsch. 23: 296-
301.
Larson, E. T., and A. L. Shanks. 1996. Consumption of marine snow by two
species of juvenile mullet and its contribution to their growth. Mar. Ecol.
Prog. Ser. 130: 19-28.
Lebour, M. V. 1922. The food of planktonic organisms. J. Mar. Biol. Assoc.
U.K 12: 644-677.
Mileikovski, S. A. 1959. Interrelations between the pelagic larvae of
Nephtys ciliata (O.F. Muller), Macoma baltica and Mya arenaria of the
White Sea. Zool. Zhurn. 38: 1889-1891.
Omori, M., and W. M. Hamner. 1982. Patchy distribution of zooplankton:
Behavior, population assessment and sampling problems. Mar. Biol. 72:
193-200.
Omori, M., and T. Ikeda. 1984. Methods in Marine Zooplankton Ecology,
John Wiley and Sons, New York, 332 pp.
Phillips, N. E., and B. Pernet. 1996. Capture oflarge particles by
suspension-feeding scaleworm larvae (Polychaeta: Polynoidae). Biol.
Bull. 191: 199-208.
Reeve, M. R. 1981. Large cod-end reservoirs as an aid to the live collection
of delicate zooplankton. Limnol. Oceanogr. 26: 577-580.
Shanks, A. L., and K. A. del Carmen. 1997. Larval polychaetes are strongly
associated with marine snow. Mar. Ecol. Prog. Ser. 154: 211-221.
Shanks, A. L., and K. Walters. 1997. Holoplankton, meroplankton, and
meiofauna associated with marine snow. Mar. Ecol. Prog. Ser. 156: 75-
86.
55
Smidt, E. L. B. 1951. Animal production in the Danish Waddensea. Meddr.
Kommn. Danm. Fisk.- og Havunders. 11(6): 151 pp.
Stommel, H. 1949. Trajectories of small bodies sinking slowly through
convection cells. J. Mar. Res. 8: 24-29.
Thorson, G. 1946. Reproduction and larval development of Danish marine
bottom invertebrates, with special reference to the planktonic larvae in
the sound (0resund). Meddr. Kommn. Danm. Fisk.- og Havunders.
Serie: Plankton 4: 7-523.
Thorson, G. 1950. Reproductive and larval ecology of marine bottom
invertebrates. Biol. Rev. 25: 1-45.
Todd, C. D., M. S. Laverack, and G.A. Boxshall. 1996. Coastal Marine
Zooplankton, second edition. Cambridge University Press, New York.
Ueda, H., A. Kuwahara, M. Tanaka, and M. Azeta. 1983. Underwater
observations on copepod swarms in temperate and subtropical waters.
Mar. Ecol. Prog. Ser. 11: 165-171.
Wilson, D. P. 1982. The larval development of three species ofMagelona
(Polychaeta) from the localities near Plymouth. J. Mar. Biol. Assoc.
U.K 62: 385-401.
Yokouchi, K. 1991. Seasonal distribution and food habits of planktonic
larvae of benthic polychaetes in Volcano Bay, Southern Hokkaido,
Japan. Ophelia 5: 401-410.
56
Bridge
Chapter III details laboratory experiments with polychaete larvae as
predators on bivalve veligers. One of these polychaete larvae, the
trochophore of the scaleworm Arctonoe vittata, is especially adept at
capturing large particles, such as veligers, and has an elaborate tuft of cilia
. .
near the mouth which may aid in engulfing large prey. Upon further
scrutiny, I found this predator-prey combination ideal for testing various
aspects of a widely-used planktonic encounter model.
One important parameter in the encounter model is the encounter
radius of the predator, or the distance at which the predator perceives prey.
Many planktonic predators detect prey by sight, vibration, or smell. Their
encounter radius can be difficult for investigators and modellers to
determine. The trochophore ofA. vittata is a contact predator. It hunts for
prey by swimming randomly, at least on a local scale, until it bumps into
prey with its anterior episphere. The encounter radius of such a contact
predator can be confidently measured as the body radius of the animal.
Other helpful attributes of this predator include continuous foraging,
relatively constant swimming speed, and a transparent gut for counting
prey.
v
In chapter IV, experiments were conducted with the trochophore of
A. vittata preying on bivalve veligers. The results of these experiments (i.e.,
numbers of bivalves consumed) were then compared to predictions based
upon encounter estimates. These comparisons allowed examination of the
1
l
importance of predator encounter radius and prey swimming speed in
calculating encounter predictions.
57
58
CHAPTER IV
THE IMPORTANCE OF ENCOUNTER RADIUS AND PREY SWIMMING
SPEED IN PLANKTONIC ENCOUNTER MODELS
Abstract
A simple predator-prey system is used to examine the importance of
encounter radius and prey swimming speed in a planktonic encounter
model. The trochophore of the scaleworm Arctonoe vittata is used as a
model predator. Comparisons were made between encounter model
estimates and actual predation observed in laboratory experiments. A.
vittata was selected because of its perpetual foraging strategy, quick prey
handling time, unambiguous encounter radius, and transparent gut.
Encounter models are sensitive to encounter radiuses and I investigate the
effects of encounter radius mis-measurements on the accuracy of
estimates. Prey swimming speed is often markedly slower than that of
cruising predators and I investigate the importance of considering slow
prey swimming speed in encounter estimations. Observations of
trochophore feeding at sul1-saturation food levels are consistent with
encounter estimates of two models: one which considers prey swimming
speed and another model which neglects it. When the predator swimming
speed is approximately one order of magnitude greater than the swimming
speed of the prey, the prey's swimming speed can be neglected without
59
significantly affecting estimates. Because of their simple foraging strategy,
planktonic polychaete larvae are a good predator system for testing
encounter models. When all parameters are carefully determined, the
planktonic encounter model can accurately estimate predator-prey
encounter rates._
Introduction
Planktonic encounter models are used to examine predator-prey
relationships in a system which is relatively inaccessible to direct
observation and experimental manipulation-the microscopic planktonic
community. Models can lend ecological meaning to observed predation
(i.e., gut contents) and may be used to estimate encounters between
planktonic predators and their prey. The first encounter model designed
explicitly for planktonic systems was that of Gerritsen and Strickler (1977).
Their 3-dimensional model, heretofore referred to as the 'GS model', has
had considerable influence on studies of plankton feeding (e.g., Giguere et
aI., 1982; Bailey and Battey, 1983; Evans, 1989; Rumrill, 1990; Luo et al.,
1996). The GS model uses encounter radius R, prey density Nh, and
predator and prey swimming speeds, v and u, respectively, to determine the
number of encounters Zpuf a single predator with its prey for v ~ u:
11: R2 Nh u2 + 3v2
Zp= 3 ( v )
Encounter radius can be difficult to measure because predators may
employ remote (e.g., visual, chemosensory, mechanical) hunting methods.
It is often difficult to measure the radius of a predator's sphere of
perception. For example, chaetognaths and some copepods hunt by
detecting the distant movements of prey with mechanoreceptors (Horridge
and Boulton, 1967; Feigenbaum and Reeve, 1977; Bailey and Yen, 1983; Yen,
1987; Yen and Nicoll, 1990; De Mott and Watson, 1991). Observations of
chaetognath feeding may overlook subtle predator responses to remotely-
sensed prey. Likewise, visual predators (e.g., Giguere and Northcote, 1987;
Giske et al., 1994) may perceive prey at greater distances than determined
by observation. Difficulty in determining encounter radius prompts an
examination of the sensitivity of the GS model's encounter estimates to
variation in the encounter radius.
As a cruising planktonic predator forages, the probability of
encountering prey increases if prey are also moving. The GS model
incorporates prey speed in estimating encounter rates. A simple
alternative model applicable to this model predator, however, can use
clearance rate to predict encounters. The clearance concept is analogous to
that presented by Rosenthal and Hempel (1970), but assumes a full
circumference of predatol'<Tadius. This clearance rate model is in many
respects similar to the GS encounter rate model, but treats prey as passive
particles, ignoring their swimming speed. In the 'CR' (=clearance rate)
model, total encounters Ep of a predator with prey are estimated using prey
61
density Nh and the total volume V of water processed (cleared) by the
predator.
For a cruising tactile encounter predator such as the trochophore of
Arctonoe vittata, the volume of water processed V is the volume ofa
cylindrical corridor searched by a predator of body radius r, over a time
period t, swimming at speed v:
V = p r 2 v t
Encounter predictions of these two models differ primarily in their
treatment of prey swimming speed. By comparing encounter predictions of
the GS vs. the CR model with actual predator encounters in a controlled
system, the importance of prey swimming speed in estimating encounters
can be determined.
In this study, I investigate the importance and utility of two
encounter model parameters: encounter radius and prey swimming speed.
I first investigate the GS model's sensitivity to encounter radius. Using the
GS and CR models, I then investigate the importance of prey swimming
speed. Comparison of models with actual encounters requires a simple
predator-prey system which satisfies basic model assumptions.
There are several factors to consider when selecting a predator to test
encounter models. The predator used in encounter model comparisons
should be a true "cruising" predator. The handling time of encountered
62
particles, prey and non-prey alike, should be negligible. The predator
should have an unambiguous encounter radius. It is also helpful if the
predators have transparent guts for convenient prey counting. Many
potential planktonic predators do not constantly cruise, but vary their speed
and even pause. These predators may be saltatory predators (O'Brien et al.,
1990), alternating periods of quiescence and movement as they forage. If
predator swimming speed is observed over a sufficiently long time, then
average overall speed might be used in calculations. Encounter estimates
for predators that swim at variable speeds may need to consider the velocity
distribution of the predator's varying swimming speeds (Evans, 1989). For
simplicity in investigating encounter models, it is best to employ a true
cruising predator swimming at a relatively constant speed. Cruising
predators may pause in foraging as they encounter both prey and non-prey
items (Hansen et al., 1991). For this reason, minimal prey handling time is
preferable for meeting the model assumptions. Many predators forage
using remote sensory perception such as visual, chemosensory, or
mechanical perception. The GS model can be most confidently applied
when the encounter radius is unambiguous and directly measurable (e.g.,
encounter radius = body radius). Finally, many potential predators are
relatively small (200-Jlm to--5-mm) and difficult to dissect. Transparent guts
enable scoring of prey items by simple observation. When a predator has
these attributes, predation experiments can reveal the strengths and
reliability of encounter models.
I used a simple predator with the attributes outlined above to
investigate the role of encounter radius and prey speed in planktonic
encounter models. The trochophore larva of the marine scaleworm
Arctonoe vittata was used as a predator upon veliger larvae of the oyster
Crassostrea gigas. The trochophore of A. vittata feeds on suspended prey,
FIGURE 1. Body radius of suspension-feeding trochophore larva of A.
vittata, synonymous with encounter radius R. Arrow indicates swimming
direction. (After Phillips and Pernet, 1996).
such as diatoms and bivalve larvae, while constantly swimming. Prey
v
handling time is negligible and a trochophore can likely capture 2 prey
items encountered only seconds apart. Trochophore foraging behavior
indicates that the animal's encounter radius is equivalent to its body
radius: foraging trochophores narrowly missing food particles do not alter
64
course or behavior. Suspended food particles are apparently not perceived
unless they actually bump into the larva's anterior episphere. Encountered
(physically contacted) food is promptly captured and ingested. The
encounter radius of A. vittata trochophores is determined by body radius
(Figure 1) and can be measured directly. Lastly, the trochophores of A.
vittata are transparent and ingested prey can be observed and counted.
Methods
Approximate 21 day-old larvae of the scale worm Arctonoe vittata
(diameter 294 mm, SD 51 mm) were used as predators and d-hinge veligers
of the oyster Crassostrea gigas, 5 to 10 days old, were used as prey. Adult
specimens of the scaleworm Arctonoe vittata, commensal with the keyhole
limpet Diodora aspera, were collected with their host from the rocky
intertidal zone on the west side of San Juan Island, Washington, USA.
Female A. vittata were spawned, eggs fertilized, and larvae cultured using
the methods described by Phillips & Pemet (1996) with the addition of
Coscinodiscus radiatus (CCMP 310) as a food source. Fertilized eggs were
cultured in 600-ml beakers at densities of - 500 P. Oyster veligers, obtained
from Whiskey Creek Oyster Farms (Tillamook, OR), were maintained in 1-
gallon jars on a diet of Isoohrysis galbana and Rhodomonas sp.
The 24 hour experiment consisted of three treatments: low, medium,
and high prey densities in filtered seawater. Each treatment was
replicated four times. One cylindrical 3-liter tank was used for each
replicate. The prey densities for the low (L), medium (M), and high (H)
65
treatments were 3, 33, and 1000 bivalve larvae 1-\ respectively. A. vittata
densities were 2 1"1 (6 tank .1), a density similar to that in the field when
trochophores were observed preying upon bivalve larvae (Johnson and
Brink, 1998). To prevent plankton from settling, all treatments and
replicates were conducted simultaneously on a roller table (Omori & Ikeda,
1984; Larson & Shanks, 1996) rolling at 1 rpm. The roller table was
maintained at 12°C in a constant temperature room at Friday Harbor
Laboratories (Friday Harbor, Washington). Experiments were run with a
14:10 light:dark cycle. Plankton do not suffer oxygen depletion during the
experimental time frame (Larson & Shanks, 1996). At the close of the
experiment, tank contents were concentrated in a partially submerged 20-
Jlm mesh nitex filter. Trochophores were quickly located and removed to
filtered seawater. Consumed bivalve veligers were then counted through
the transparent gut of each trochophore.
The number of prey ingested by each predator were then used to
examine the sensitivity of the GS model to encounter radius and the
importance of prey swimming speed. For the examination of encounter
radius, the number of prey ingested was compared to that predicted by the
GS model using an encounter radiuses of 210 Jlm (A. vittata's body radius
and actual encounter radiRS), 420 Jlm, and 1000 Jlm. For the investigation
of prey swimming speed, prediction curves were generated based on
different relative magnitudes of predator and prey swimming speeds.
Predator swimming speed used in calculations was 2.56 mm S·l (SD = 0.53,
Pernet, unpublished). Prey swimming speed used was 0.30 mm s'\
66
consistent with observed swimming speeds of C. gigas d-hinge veligers
(pers. obs.) and published horizontal swimming speeds of d-hinge
Crassostrea virginica (Hidu and Haskin, 1978). Theoretical curves
generated by the model were then compared to the actual data to determine
the importance of prey swimming speed in estimating encounter rates.
Results
Encounter Radius Sensitivity
Figure 2 superimposes the mean number of prey gut-Ion three
encounter probability curves predicted by the GS model, each calculated
using different encounter radiuses: 210 Jlm (A. vittata's mean body radius
and actual encounter radius), 420 Jlm, and 1000 Jlm. Assuming a capture
occurs with each encounter, the number of veligers consumed by individual
trochophores is consistent with the predictions of the GS model in the Low
and Medium prey density treatments. In the High prey density treatment,
however, the number of veligers consumed is less than predicted by the
models. This is probably a result of predator saturation at high prey
densities (see Discussion).
'Prey Swimming Speed
The predicted curves from the GS and CR models are plotted in
Figure 3. The curves overlap completely, giving the appearance of a single
curve. Figure 3 also superimposes predation data on the prediction curves.
Assuming a capture occurs with each encounter, then the number of
encounters by the trochophores at the Low and Medium prey densities is
12 R=0.21mm
,-..
.....
.~
R=0.42mm~
-"'C 10
.....
~
-& 8
'**'
'-"
'"d
CD 6S
~ I00~ 40Co)~
CD 2$-I
~
0
0 0.5 1 1.5 2 2.5 3 3.5
log Prey Density
FIGURE 2. The number of captured prey in Arctonoe vittata trochophores
vs.log prey density. The black curve is the GS prediction using actual
encounter radius (=body radius of 210 JlDl) of the predator A. vittata. The
checkered curve is the prediction when the radius is 420 JlDl. The gray
curve is the prediction when the radius is 1000 JlDl. Black dots indicate the
mean # of prey gut-1 at each prey density (n=24 trochophores for each prey
density). Error bars are 95% Confidence Intervals.
consistent with the predictions from both models. In the High prey density
""treatment, however, the numbers of veliger larvae consumed is less than
predicted by the models. As previously mentioned, this may be explained by
predator satiation at the highest prey density (see Discussion).
68
:;--. 12
.~
oj
::' 10
-
GS prediction
~
So XXXXX)l CR prediction
:t:t:: 8
'-"
"C(J) 6S
~ I00~ 40U~ 2
~
0
0 0.5 1 1.5 2 2.5 3 3.5
log Prey Density
FIGURE 3. The number of prey in Arctonoe vittata trochophores vs. log
prey density. The black curve is the prediction of the GS model. The thin
white curve, directly overlaying the black curve, is the CR model prediction.
Black dots indicate the mean # of prey gut-1 at each prey density (n=24
trochophores for each prey density). Error bars are 95% Confidence
Intervals.
Discussion
The number of veligers consumed by trochophores at the Low and
Medium prey densities is consistent with the predictions of the GS model.
It appears the model accurately predicts the number of bivalves eaten by A.
vittata larvae. It should be noted that, although capture success may not
always be 100%, it is quite"high and predation variance may then obscure
potential discord between predation observations and encounter
predictions.
Encounter Radius
When the modeled encounter radius was increased, observed
encounters no longer fell directly on the curves (Figure 2). Larger-than-
actual encounter radiuses fail to predict the number of veligers ingested.
The encounter radiuses, used to help create the range of prediction curves,
are intended to reflect potential mistakes in the measurements of
ambiguous remote encounter radiuses. A radius of 420 Jlm may seem far
offA . vittata's actual encounter radius of 210 Jlm. Indeed, it exceeds the
trochophore body radius by 100%. For small predators with ambiguous
remote-sensing encounter radiuses, however, a slight anthropogenic
mistake in radius determination could easily result in 100% overestimation
of actual encounter radiuses. The GS encounter model is sensitive to the
encounter radius parameter. Careful observations are needed to reliably
determine encounter radiuses for use in encounter rate model estimates.
With High prey density, the actual number ofveligers consumed by
A. vittata larvae is less than predicted by models. This is most likely a
result of predator satiation. Indeed, A. vittata larvae in the High prey
density treatment, with an average of 4 veligers each, appeared to have very
full guts. Predators from the High prey density treatment did, however,
have a wide range ofveliger numbers in their guts (0-10). One might expect
less variation in predation"if encounters far exceed consumption ability.
This variation might be explained, however, by the relationship between
number of prey per gut and predator size: larger trochophores ingest more
veligers. The trochophores used in this experiment, despite the fact that
they were of equal age, varied in size. Predator size significantly correlated
70
with the number of prey captured (r =.79; df= 1,22; p = 0.000005). This
suggests that predation upon veligers in the High prey density treatment
was limited more by predator capacity than prey availability or encounters.
Holling (1959) describes a type II predator functional response curve,
where predation rate decreases as predator satiation sets an upper limit to
food consumption. The type III curve resembles the type II curve in having
an upper limit to consumption, but differs in that low consumption at low
prey density results from the lack of a search image or low hunting
efficiency. Low consumption described by a type II curve, on the other
hand, is simply the result of few encounters. If we assume the models
accurately predict encounters, the agreement between model predictions
and actual consumption at the Low and Medium prey densities is
consistent with a type II curve. This in turn supports the claim that A.
vittata trochophores forage by cruising randomly and have a high veliger
capture success rate.
Comparison of actual predation by A. vittata trochophores with
model predictions reveals that estimates are sensitive to increased
encounter radiuses. Even estimates calculated using an encounter radius
100% too large are inaccurate and, therefore, it is important to reliably and
accurately determine encounter radiuses.
Prey Swimming Speed
The main difference between the two encounter models is that the CR
model treats prey as passive particles while the as model does not. When
prey swimming speed is low the difference between model outputs small. If
71
FIGURE 4. Encounter prediction curves for the GS and CR encounter
models: curves 1 & 2 (GS and CR, respectively) when predator and prey
swimming speeds (v and u) are both low (C. gigas veliger swimming speed
used for both predator and prey); curves 3 and 4 (GS and CR, respectively)
when v and u are both high (A. vittata trochophore swimming speed used
for both predator and prey); curves 5 and 6 (GS and CR, respectively) with
appropriate v and u values for A. vittata and C. gigas (superimposed curves
plotted from Figure 3).
2.92.7
2. CRu &vlow
1. GS u &vlow
3. GS u & v high
4. CR u & v high
5. GS v high, u low
-
6. CR v high, u low
1.7 1.9 2.1 2.3 2.5
log Prey Density
prey speed is very low relative to that of the predator, then prey speed is
negligible with regard to estimating encounters (see Figure 4, curves 5 and
6). When predator and prey swimming speeds are more similar than those
ofA. vittata and C. gigas, prey speed may affect encounter estimates and
should be considered in calculating encounter estimates. To illustrate this,
I used more similar predator and prey swimming speeds to estimate
encounters and then compared between the GS and CR models. Figure 4
illustrates GS and CR prediction curves for three relative predator and prey
72
swimming speeds. Model comparisons are made at low similar speeds
(curves 1 and 2), high similar speeds (curves 3 and 4), and disparate speeds
(curves 5 and 6). For the low swimming speeds, the swimming speed of
bivalve veligers (0.03 cm S-l) was always used. For high swimming speeds,
the swimming speed of A. vittata (0.26 cm S-l) was always used. For
example, 0.30 mm S-l was used as the swi~ming speed for both the
predator and prey in calculating prediction curves 1 and 2. For every set of
comparisons, one curve is the prediction of the GS model (curves 1, 3, and 5)
and the other curve is the prediction of the CR model (curves 2, 4, and 6).
Unlike the nearly identical prediction curves resulting from calculations
with disparate swimming speeds, The GS and CR prediction curves,
overlaying one another when predator and prey swimming speeds are
similar, separate as swimming speeds diverge. In other words, when
swimming speeds (u and v) become similar, whether low or high, the
prey's swimming speed becomes important in calculating encounter
predictions. This phenomenon is more pronounced (i.e., the GS and CR
curves are farthest apart) when similar swimming speeds are also high
(Figure 4, curves 3 and 4). The swimming speed ofA. vittata trochophores
is approximately one order of magnitude greater than the swimming speed
of C. gigas veligers. v'
Like the larvae ofA. vittata, many predators swim faster than their
prey. In these cases, both the GS and CR models may be employed with
equal utility. One need only account for prey swimming speed, or use the
GS model, when predator and prey swimming speeds are relatively
73
similar. How low relative to predator swimming speed must prey
swimming speed be before it can be ignored? In A. vittata and C. gigas,
where prey speed was negligible in estimating encounters, predator
swimming speed exceeds that of the prey by approximately an order of
magnitude. One order of magnitude or greater relative difference in
predator and prey swimming speeds may be an appropriate cut-off for
neglecting prey speed in encounter estimates-a cruising predator is
primarily responsible for prey encounters when its swimming speed is lOx
that of its prey.
Conclusion
Predation rates decrease when invertebrate larvae are presented to
predators at near-natural densities which are relatively low. Consequently,
predation on planktonic invertebrate larvae has rarely been observed under
the most natural laboratory conditions (Johnson and Shanks, 1997;
Johnson and Brink, 1998). The results of this study's comparisons between
predation and encounter model predictions indicate that low or absent
predation may be explained in part by low encounters at near-natural prey
densities. Johnson and Brink (1998) examined predation by Arctonoe
vittata trochophores on Crf!ssostrea gigas veligers with the same methods,
range of prey densities, and results as the current study. Likewise,
Johnson and Shanks (1997) observed low or absent predation for 4 predator-
prey combinations when larvae were presented at near-natural prey
densities. Predator-prey combinations included Upogebia pugettensis zoeae
74
feeding on urchin blastulae, U. pugettensis zoeae feeding on urchin plutei,
unidentified hydromedusae feeding on barnacle nauplii, and Obelia sp.
hydromedusae feeding on urchin blastulae. These predators may not
regularly consume these larvae at natural densities, which are low. The
natural diet of these planktonic, suspension-feeding predators may be
. .
frequently encountered phytoplankton, protists, or abundant metazoans,
rather than relatively scarce invertebrate larvae. If this is the case,
invertebrate larvae may be rarely consumed because they are rarely
contacted.
The Gerritsen and Strickler (1977) planktonic encounter model can
accurately predict encounters when details of predator-prey interactions
are known. These details include the ratio of the predator's time spent
foraging, the predator's average swimming speed when foraging, the
relative prey swimming speed in relation to the predator, prey density, and
predator encounter radius. The most difficult parameter to measure in
remote-sensing predators may be the encounter radius, but the radius
must be determined accurately because the GS model is sensitive to radius
variation. Prey swimming speed can usually be determined without
difficulty, but may be unnecessary to include when prey are slow relative to
the predator. In this latter'case, the GS and CR models may be employed
with equal utility. Encounter estimates enhance investigations of
population biology, planktonic mortality, and behavioral ecology. Accurate
measurements of encounter radius and knowledge of when to include prey
75
swimming speed are important for correct estimation of planktonic
encounters.
Selected References
Bailey,K.M. and Battey, R.S. (1983) A laboratory study of predation by
Aurelia aurita on larval herring, Clupea harengus: experimental
observations compared with model predictions. Mar. Biol., 72, 295-302.
Bailey,K.M. and Yen,J. (1983) Predation by a carnivorous marine copepod,
Euchaeta elongata Esterly, on eggs and larvae of the Pacific hake,
Merluccius productus. J. Plankton Res., 5,71-82.
DeMott,W.R. and Watson,M.D. (1991) Remote detection of algae by
copepods: Responses to algal size, odors and motility. J. Plankton Res.,
13, 1203-1222.
Evans,G.T. (1989) The encounter speed of moving predator and prey. J.
Plankton Res., 11,415-417.
Feigenbaum,D. and Reeve,M.R. (1977) Prey detection in the Chaetognatha:
response to a vibrating probe and experimental determination of attack
distance in large aquaria. Limnol. Oceanogr., 22, 1052-1058.
Gerritsen,J. and Strickler,J.R. (1977) Encounter probabilities and
community structure in zooplankton: a mathematical model. J. Fish.
Res. Board Can., 34, 73-82.
Giguere,L.A., Delage,A., Dill,L.M., and Gerritsen,J. (1982) Predicting
Encounter Rates for Zooplankton: A Model Assuming a Cylindrical
Encounter Field. Can. J. Fish. Aquat. Sci., 39,237-242.
Giguere,L.A. and Northcote,T.G. (1987) Ingested prey increase risks of
visual predation in transparent Chaoborus larvae. Oecologia, 73, 48-52.
Giske,J., Aksnes,D.L., and'Fiksen,O. (1994) Visual predators,
environmental variables and zooplankton mortality risk. Vie Milieu, 44,
1-9.
Hansen,B., Hansen,P.J., and Nielsen,T.G. (1991) Effects of large
nongrazable particles on clearance and swimming behaviour of
zooplankton. J. Exp. Mar. Biol. Ecol., 153,257-269.
76
Hidu,H. and Haskin,H.H. (1978) Swimming speeds of oyster larvae
Crassostrea virginica in different salinities and temperatures.
Estuaries, 1, 252-255.
Holling,C.S. (1959) The components of predation as revealed by a study of
small mammal predation of the European pine sawfly. Canad. Entomol.,
91, 293-320.
Horridge,G.A. and Boulton,P.S. (1967) Prey detection by Chaetognatha via a
.vibration sense. Proc. R. Soc. Lond., Ser. B., H~8, 413-419.
Johnson,K.B. and Shanks,A.L. (1997) The importance of prey densities and
background plankton in studies of predation on invertebrate larvae. Mar.
Ecol. Prog. Ser., 158,293-296.
Johnson,K.B. and Brink,L.A. (1998) Predation on bivalve veligers by
polychaete larvae. Biol. Bull., In press.
Larson,E.T. and Shanks,A.L. (1996) Consumption of marine snow by two
species of juvenile mullet and its contribution to their growth. Mar. Ecol.
Prog. Ser., 130, 19-28.
Luo,J., Brandt,S.A. and Klebasko,M.J. (1996) Virtual reality of
planktivores: a fish's perspective of prey size selection. Mar. Ecol. Prog.
Ser., 140, 271-283.
O'Brien,W.J., Browman,H.I. and Evans,B.I. (1990) Search strategies of
foraging animals. American Scientist, 78, 152-160.
Omori,M. and Ikeda,T. (1984) Methods in Marine Zooplankton Ecology,
John Wiley & Sons.
Phillips,N.E. and Pemet,B. (1996) Capture of large particles by suspension-
feeding scaleworm larvae (Polychaeta: Polynoidae). Biol. Bull., 191, 199-
208.
Rosenthal,H. and Hempel,G. (1970) Experimental studies in feeding and
food requirements of herring larvae (Clupea harengus L.). In Steele,J.H.
(ed), Marine Food Chain"S. Oliver and Boyd, Edinburgh, pp. 344-364.
Rumrill,S.S. (1990) Natural mortality of marine invertebrate larvae.
Ophelia, 32, 163-198.
Yen,J. (1987) Predation by a carnivorous marine copepod, Euchaeta
norvegica Boeck, on eggs and larvae of the North Atlantic cod Gadus
morhua L. J. Exp. Mar. Biol. Ecol., 112, 283-296.
71
Yen,J. and Nicoll,N.T. (1990) Setal array on the first antennae of a
carnivorous marine copepod, Euchaeta norvegica. J. Crust. Bioi., 10, 218-
224.
78
Bridge
The most valuable data in this doctoral dissertation is that acquired
in field observations and experiments in chapter V. Field observations
measure predation on a variety of invertebrate larvae by a natural plankton
assemblage. Observed predation is comparE:ld to estimated encounters of
larvae with predators in corrals using the planktonic encounter model
examined in chapter IV. Data from field experiments manipulating prey
density and background plankton, analogous to laboratory experiments
described in chapters II and III, are also given in chapter V. Generally,
the results of field experiments corroborate findings in the laboratory-
predation is low when larvae are presented at natural densities with
background plankton present.
79
CHAPTER V
PREDATION ON PLANKTONIC MARINE AND ESTUARINE
INVERTEBRATE LARVAE
Abstract
Predation on invertebrate larvae is a potentially important source of
mortality and may influence the numbers of larvae in the plankton. We
conducted in situ and laboratory experiments to evaluate predation on
larvae by near-natural plankton assemblages. Corrals, or mesocosms,
were used to capture a column of whole seawater and its inherent suite of
potential predators at natural densities. To these corrals, we added known
numbers of larvae marked with calcein. Marked larvae included the plutei
of the sand dollar Dendraster excentricus, veligers of the oyster Crassostrea
gigas, and veligers of the snail Littorina plana. Most potential predators
were captured with the natural plankton assemblage, but some were added
to corrals at natural densities to ensure their universal inclusion in
experiments. A wide variety of potential predators were captured in or
added to corrals. After 24-h corrals were collected and the fate of marked
larvae, determined. Predation on unmarked invertebrate larvae, captured
at natural densities as background plankton, was also quantified. Recovery
of marked larvae was often 100%, enabling a thorough and direct
determination of predators and predation rates. Under the most natural
conditions, only 1 of 9 experiments with plutei showed any pluteus
predation (a single individual in the gut of the hydromedusa
Proboscidactyla flavicirrata). Veligers were also often completely
untouched, but appear to experience more predation than plutei. The main
predators of bivalve veligers were the heterotrophic dinoflagellate Noctiluca
. .
scintillans and the hydromedusa Proboscidactyla flavicirrata. Potential
instantaneous mortality due to N. scintillans was as high as -.07 day·l and
could result in a loss of 87% of a planktonic population in 28 days.
To learn causes of the overall low predation, in situ corral
experiments were conducted which presented marked prey at near-natural
and unnaturally high prey densities and in the presence or absence of
natural background plankton (i.e., the natural suite of diverse plankton in
whole, unfiltered seawater). All predation observed at high prey densities
decreased or disappeared at near-natural prey densities. This suggests
that part of the lack of predation in corrals could be the result of few
predator-prey encounters. Predation was also decreased by the presence of
natural background plankton. Background plankton may occupy the
predators time and decrease opportunities for encounters with larvae,
obscure larvae from detection or capture, or serve as substitute food. The
impact of predation on corr~l populations was extremely variable, which is
probably the case in the field. While N. scintillans was a significant
predator in 2 of 10 in situ experiments, the majority of observations showed
that veliger and pluteus larvae suffered little or no predation. Thus,
planktonic predation may not always be a major source of larval mortality.
81
Introduction
The majority of marine and estuarine invertebrates have complex
life-histories that include planktonic larvae which reside in the water
column for hours to months (Levin and Bridges, 1995). Larvae may develop
from free-spawned ova or be released fro~ adults or egg cases after a period
of brooding or encapsulation. A single adult invertebrate can produce vast
numbers of these planktonic propagules. For example, during a single
spawning season one female sand dollar (Dendraster excentricus) can
spawn 3.8 x 105 eggs (Morris et aI., 1980), one female dungeness crab
(Cancer magister) can release 2.5 x 106 larvae (Morris et aI., 1980), a female
oyster (Crassostrea gigas) can spawn 55.8 x 106 ova (Galtsoff, 1964), and the
sunflower star Pycnopodia helianthoides may release as many as 160 x 106
eggs (Chia and Walker,1991). These larvae then develop in the plankton
until competent for settlement and metamorphosis. The numbers of
competent larvae present in local plankton correlates with recruitment to
benthic communities (Connell, 1985; Gaines et aI., 1985; Roughgarden et
aI., 1991). These studies of supply-side ecology have investigated the
important relationship between planktonic larval supply and benthic
community composition. Recruitment to benthic populations can be
determined by the supply of larvae available in the plankton (Roughgarden
et aI., 1984; Connell, 1985; Gaines et aI., 1985; Roughgarden et aI., 1991).
The number of new recruits to the benthic adult assemblages can be
high-the barnacle Semibalanus balanoides was observed to settle in
82
densities reaching 215 individuals cm-2 (Connell, 1985). When compared to
the area's estimated propagule production, however, these newly
metamorphosed juveniles are few. Some studies attempt to estimate
mortality rates by contrasting propagule production with benthic
recruitment (see Rumrill, 1990 for review). These studies have estimated
mortality rates to be from 0.03 day-l in the cone snail Conus quercinus
(Perron, 1986) to 0.80 day-l in the clam Mya arenaria (Ayers, 1956), but
cannot distinguish between larval and early juvenile mortality. Many
planktonic mortality studies suffer from the drawbacks and potential biases
of anecdotal information and indirect evidence (Strathmann, 1985), limiting
available reliable knowledge of the sources and importance of mortality.
High mortality rates are expected, however, because invertebrate
populations are generally stable over time, and mortality must occur
between spawning or release and recruitment. Possible sources of
planktonic mortality include fertilization failure (Babcock et aI., 1992;
Brazeau and Lasker, 1992), starvation (Olson and Olson, 1989; Boidron-
Metairon, 1995), lethal temperatures (Pechenik, 1987), the absence of the
proper settlement substratum (Jackson and Strathmann, 1981), transport
away from suitable settlement sites (Farrell et aI., 1991; Roughgarden et aI.,
1991), predation on embryos and larvae (Rumrill, 1990; Morgan, 1995a), and
pathogens and genetic abnormalities (Thorson, 1946, 1950, 1966; Rumrill,
1990). Only pathogens and genetic abnormalities have not been
investigated.
83
Planktonic invertebrate larvae can be consumed either by benthic
suspension-feeders or planktonic predators. Planktonic embryos and
larvae may encounter benthic suspension-feeding predators shortly after
release, incidentally during their planktonic life, or as they attempt to settle
and test the benthos for a suitable substratum. Benthic predators may form
a "wall of mouths" (Emery, 1973) and can make the acquisition of a
settlement site a hazardous undertaking. Organisms associated with coral
reefs may consume as much as 60% of passing zooplankton, which
included the larvae of crustaceans, polychaetes, cnidarians, molluscs, and
echinoderms (Glynn, 1973). Suspension-feeding barnacles also inhibited
recruitment of colonial ascidians and bryozoans in field experiments
conducted by Young and Gotelli (1988). The anthozoans Alcyonium
siderium and Metridium senile captured and consumed planktonic
invertebrate larvae (Sebens and Koehl, 1984). Not all suspension-feeders,
however, consume invertebrate larvae. For instance, Bingham and
Walters (1989) found that settling larvae escaped predation by suspension-
feeding ascidians and Rumrill (1987) calculated the risk of predation, by 2
species of benthic suspension-feeders consuming Asterina miniata
brachiolaria larvae, to be 1.2% per saltation event (i.e., settlement or re-
suspension). Additional e~dence of low predation by ascidians includes
their lack of effect on larval recruitment in a study by Young (1989).
Mortality of settling larvae by benthic suspension-feeders is clearly variable,
but much more investigation is necessary to determine the overall risk of
predation presented by benthic suspension-feeders.
84
Planktonic predators of invertebrate larvae have been studied in the
laboratory, in the field through correlation of high predator abundance and
larval decline, and by gut content analysis of field-caught predators.
Laboratory experiments have investigated the following factors and their
effect on larval predation rates: antipredator defenses (Pennington and
. .
Chia, 1984; Morgan 1987, 1989), developmental stage and post-contact
behavioral responses (Rumrill et aI., 1985; Pennington et aI., 1986), larval
size (Pennington and Chia, 1984; Rumrill et aI., 1985; Rumrill, 1987),
container size (Toonen and Chia, 1993), prey density (Rumrill et aI., 1985;
Pennington et aI., 1986; Johnson and Shanks, 1997, Johnson and Brink,
1998), and background plankton presence (Johnson and Shanks, 1997;
Johnson and Brink, 1998). In a review on larval mortality, however,
Rumrill (1990) points out a caution with regard to laboratory experiments
on predation:
An important limitation is that the majority of laboratory experiments
have been conducted in small containers at prey densities that are 2 to
3 orders of magnitude greater than natural densities of larvae in the
plankton. Direct extrapolation of mortality rates from laboratory
studies is unwarranted because rates of predation in the laboratory
are strongly dependent upon the size of the experimental container.
(Rumrill, 1990, p. 173)
In order for laboratory experiments to provide information that is directly
applicable to estimates of natural mortality, much more information must
first be collected about specific natural predator-prey relationships with
confirmation that the containers employed do not create artifacts.
85
Predation on planktonic larvae can be studied in the field by
identifying pelagic predators whose abundance is inversely correlated with
that of larvae. One example of this is the predatory ctenophore Mnemiopsis
leidyi, whose abundance has been negatively correlated with larval
abundance and recruitment of crustaceans and fish (Nelson, 1925; Burrell
aiJ.d Van Engel, 1976; Cowan et aI., 1994). This 'method requires, however,
the fortuitous monitoring of key predators and the need to assume that
larval decline in the plankton is due, in whole or part, to predation by these
predators.
Another method of monitoring predation on planktonic invertebrate
larvae is by gut content analysis of potential predators. Indeed, predators
have been identified based upon their gut contents. Examples of
invertebrate larval predators identified in this manner include the
ctenophore Mnemiopsis leidyi (Nelson, 1925; Burrell and Van Engel, 1976),
the hydromedusa Phialidium -sp. (McCormick, 1969), decapod larvae
(Lebour, 1922), and salmon fry (Bailey et aI., 1975). Unfortunately, some of
these predators may have consumed larvae in cod-end plankton buckets
and predation may be an artifact of collection. For example, chaetognaths
are known to feed unnaturally or at increased rates on plankton in
collection reservoirs (Feigel1baum and Maris, 1984).
Because predation in the plankton may be determined by opportunity,
or encounters between predators and prey (see laboratory evidence of
density-dependent predation- Rumrill et aI., 1985; Pennington et aI., 1986;
Johnson and Shanks, 1997, Johnson and Brink, 1998), many predators may
86
feed unnaturally when concentrated with potential prey in plankton
samples. Even if it is assumed, however, that the presence of larvae in
predator guts is not an artifact of collection, predator gut analysis has not
been an effective method for evaluating the impact of predation on larval
populations. When gut content studies have identified predators, the focus
. .
has often been on the composition of the predator's diet. Invertebrate larvae
are a minor part of predator diets. However, the relative importance of
predation has not yet been determined for an individual larva throughout
the duration of larval life. Since data on the concurrent density of
planktonic prey is rarely offered, little can be said about the importance of
particular predators in the ecology of larval invertebrates. For example,
Bailey et al. (1975) observed that salmon fry had preyed upon decapod
larvae. Only 9% of salmon fry guts sampled, however, contained decapod
larvae. Decapods only represented 1% of the diet by volume. Decapod
larvae are not likely to be an important component of young salmon diets
and nothing is known of the potential impact on decapod populations by
salmon predators. Some hydromedusae of Phialidium sp. consume
invertebrate larvae, but less than 10% of predators sampled contained
larvae and, in those, larvae comprised less than 3% of identified prey
(McCormick, 1969). As with decapod larvae and salmon fry, these larvae
are not likely to be an important component ofPhialidium sp. diets and
nothing is known of the potential impact on larval populations by this
predator. Because this data focuses on predator diets rather than larval
risk, important questions still remain. How important is planktonic
predation over a larva's planktonic life? What is the daily risk of predation
for an individual larva from all potential predators?
It is possible to evaluate predation risk using gut contents in
combination with known digestion rates and field densities of predators and
prey. This has been done to evaluate predation on adult copepods by larval
fish (Purcell, 1990) and on copepods, fish eggs, and fish larvae by
coelenterates (Purcell et aI., 1994; Chandy and Greene, 1995). These results
cannot be extrapolated to predation on invertebrate larvae because these
predators may preferentially consume copepods, larval fish, or eggs, and
digestion times vary with prey type and size (Purcell, 1982; Chandy and
Greene, 1995). According to Purcell (1982), this combination approach to
evaluating in situ predation requires accuracy in measurements of
digestion times for particular prey, identification of digested prey,
converting size to dry weight and carbon, and determining predator and
prey densities from plankton tows. We would add that predator digestion
times for particular food types can vary tremendously depending on the
total amount of food in the gut. For instance, trochophore larvae of the
scaleworm Arctonoe vittata will pass bivalve veligers within 3 to 4 hours
when several veligers have been consumed and more are available. A lone
veliger in the gut ofA. vittata, however, may remain in the gut for as long
as a day (K. Johnson, pers. obs.). In spite of the potentially inaccurate
assumptions, estimates using gut contents, digestion times, and densities
may more accurately estimate field mortality than estimates based upon
laboratory predation studies (Purcell, 1982). Laboratory studies of predation
88
are potentially fraught with behavioral artifacts (Reeve 1977, 1980), but it is
unknown whether indirect field studies or laboratory experiments provide
the best estimate of field mortality.
This study describes experiments conducted both in the field and in
the laboratory. All experiments examined predation on larvae under the
most natural conditions possible. Most field studies were designed to
simply observe predation on invertebrate larvae, expose predator identities,
and determine predation rates under near-natural conditions. These
observational field studies test the hypothesis that populations of
invertebrate larvae suffer significant predation in near-natural plankton
assemblages. To examine factors affecting predation rates, field and
laboratory studies test the hypotheses that 1) proportion of predation on a
larval population changes with prey density and 2) natural background
plankton reduces predation rates.
This study uses natural assemblages, including a diverse suite of
potential predators, enabling us to directly determine the predation risk for
experimental larval populations. Corrals were inoculated with marked
and enumerated invertebrate larvae at the start of 24 h experiments. By
marking prey, we could know initial prey densities, retrieve larvae after the
experiment, determine the n:umber of survivors, and identify the natural
predators. Observations of predation are direct and can be related directly
to the potential impact of predation on experimental populations of
invertebrate larvae. Corral assemblages also included wild (i.e., randomly
caught and unmarked) invertebrate larvae at natural densities. We were
89
able to examine predation risk for captured wild larvae using predator gut
content analyses, known wild prey densities, and a planktonic predator-
prey encounter model. Finally, corrals were also used to manipulate prey
density and "background plankton" presence, examining their effect on
predation rates.
These field observations and experiments provide a direct
assessment of predation's importance for larvae in an experimental
assemblage. The following strengths of experimental design contribute to
the data's value for examining predation as an important source of larval
invertebrate mortality. Larval species examined as prey include 3 species
of marked larvae and several other species of wild larvae representing 4
phyla. The data were collected from interactions with captured natural
assemblages, each with a diversity of potential predators representing all
defined planktonic feeding strategies (Greene, 1985). Initial densities of
marked larvae were known, allowing more powerful analysis of
observations. Predators and prey interacted at natural densities,
eliminating the possibility of artifactual behavior induced by abnormal
densities. Marked larvae were easily visible in the guts of predators.
Corral volumes far exceeded the practical capacity of laboratory containers
and reduced artifacts resulting from small volumes. Finally, corral
samples were collected and fixed immediately at the close of experiments,
minimizing the possibility of artifactual predation in the concentrated
sample. Studies such as this, investigating in situ predation on larvae, can
be valuable for determining the potential sources of significant larval
00
mortality, analyzing factors that shape benthic communities, discussing
the life history evolution of benthic marine invertebrates with complex life
cycles, and identifying critical areas of future research.
Methods
Marked Larvae
Marked larvae included pluteus larvae of the sand dollar Dendraster
excentricus, veliger larvae of the snail Littorina scutulata, and d-hinge
veliger larvae of the oyster Crassostrea gigas. Adult breeding stock of D.
excentricus were collected from the North Spit, Coos Bay, Oregon and from
West Sound, Orcas Island, Washington. Adult L. scutulata were collected
from Sunset Bay, Oregon and Friday Harbor, Washington. Embryos and
larvae were obtained using methods described in M. Strathmann (1987). D-
hinge larvae of the oyster C. gigas were provided by Whiskey Creek Oyster
Farms (Tillamook, Oregon).
Larvae were marked with fluorescent Calcein (Sigma Corp.), which
is permanently incorporated into skeletons as calcium carbonate is laid
down. Larvae were cultured in filtered seawater on diets of Isochrysis
galbana and Rhodomonas sp. in the presence of Calcein at a concentration
v
of 200-500 ppm. Larval behavior and development appears normal in the
presence of Calcein (Rowley, 1993; pers. obs.). Calcein is primarily
comprised of CaCOs, which should not affect predator preference for larvae.
Though there is also a fluorescent molecule in Calcein, it is bound in the
91
larval skeleton. Even if the fluorescene does taste bitter to the predator,
lethal predation would be required for the unpleasant taste to be discovered.
For example, the Calcein incorporated into sand dollar skeletons is
completely internal and would not be tasted until after the maceration of
dermal tissue. Finally, digested skeletons retain their Calcein component
and are visible in the guts of predators and in fecal peilets.
Seeded Predators
In some experiments (see Tables 1 and 2) naturally occurring
predators were seeded in corrals at natural densities to ensure consistent
representation in all treatments and replicates. Seeded predators were
collected at high tide by slowly towing a plankton net equipped with a large
blind cod-end (after Reeve, 1981). Predators were quickly removed from the
plankton sample to isolation in filtered seawater using a large bore pipette
or a small cup. Experiments began within 24 hours of predator collection
One predator, the trochophore larva of the scaleworm Arctonoe
vittata, was cultured for use in corrals. Adult specimens, commensal with
the keyhole limpet Diodora aspera, were collected with their host from the
west shore rocky intertidal of San Juan Island, Washington. Individuals of
A. vittata were spawned atrd larvae were cultured using the methods
described by Phillips and Pernet (1996) with the addition of Coscinodiscus
radiatus (CCMP 310) as a food source. Fertilized eggs were cultured in 600-
ml beakers at densities of -500 t 1• Larvae approximately 21 days old were
used as predators in experiments.
92
Corrals, Deployment, and Collection
Experiments were conducted in corrals made of flexible 20-mil clear
PVC sheeting. Corrals held 123 liters of seawater when deployed. For
corral design and dimensions, see Figure ID. Corrals were water tight
. .
except for cod-end collection buckets. Each bucket had 8 portholes covered
with 53-Jlm Nitex mesh and a total filtering area of 176.4 cm2• For
deployment, corrals were collapsed longitudinally and fastened in the
collapsed position with a securing line (Figure lA). Corrals were
submerged and lowered to the appropriate depth (see deployment depth,
Figure lA) with a 3-point bridle, harness and line. After the corral was at
depth and the disturbed water column had cleared from directly above the
corral, the securing line was released (Figure IB). The corral was then
drawn slowly surfaceward with the bridle line (Figure lC). The cod-end
bucket remained at depth as the mouth of the corral was drawn
surfaceward, eventually breaking the surface of the water. The corral was
suspended by floats for the 24 h experiment. This resulted in the quiet
capture and isolation of a natural assemblage of plankton and potential
predators (See corral contents, Table 5), including delicate predators such
as chaetognaths and coelen1erates, at natural densities.
It should not automatically be assumed that behavior and feeding in
corrals reflects that which would occur in a natural environment. The
feeding behavior of some predators can be altered by small laboratory
containers (e.g., Toonen and Chia, 1993) and natural turbulence,
dampened in corrals, enhances encounters between planktonic organisms.
The most compelling argument for low container effects in corrals is the
fact that large numbers of predators fed on larvae under unnatural
laboratory conditions, but did not consume them in corrals. It is most
parsimonious to assume that, as conditions were made more natural (i.e.,
increased container volume and turbulence relative to the laboratory),
predation in 123-1 corrals more closely reflects nature than predation
observed in 3-1 laboratory containers.
At the close of experiments, corrals were hauled from the sea and
captured water exited through the cod-end bucket. Contents were fixed
immediately in 4% buffered formalin. Filter screens were washed
repeatedly to free plankton from the mesh and ensure collection of all of the
sample. Samples were stored in the dark.
Observation Experiments
One observation experiment consisted of 4 corrals inoculated with
known numbers of marked larvae. A total of 10 experiments were
conducted, 2 in the boatbasin at Charleston, Oregon (43°21'10" N, 124°19'50"
W), and 8 from the dock of Friday Harbor Laboratories, Friday Harbor,
Washington (48°32'10"N, 1~3°00'19"W). In some experiments, rare
suspected potential predators were added to corrals. These potential
predators were present in surrounding plankton, but were seeded in
corrals to ensure their consistent representation in the randomly captured
assemblages. Table 1 provides a summary of experiment location,
FIGURE 1. In situ corrals: deployment and dimensions. A. Corral with
slack line securing the collapsed position is lowered to depth with taught
bridle line. B. At depth, slack line unties as it is pulled, freeing the corral
for expansion. C. With the bridle line, the corral is pulled surfaceward,
quietly capturing a column of seawater and natural plankton assemblage.
D. Corral is suspended at the sea surface by floats for the duration of the
experiment. Corral volume is approximately 123 liters.
30 em
~
numbers of marked larvae added, and the identities and numbers of seeded
potential predators. Table 1 also indicates whether experiments were
started during the day or night, which should effect which potential
predators are randomly captured in corrals. Night experiments were
started between 10:00 pm and 1:00 am.
After corrals were deployed, a known number of marked larvae were
added to each corral. Marked larva corral densities were intended to reflect
the high end of potential natural densities. Our survey of the literature
indicates that, for many invertebrate larvae, the experimental densities
selected (0.4 to 1.0 liter-I) are reasonable natural densities (Carriker, 1951;
Zimmerman, 1972; Cameron and Rumrill, 1982; Rumrill et aI., 1985;
Emlet, 1986; Miller, 1995). Following inoculation, corral assemblages were
mixed with 2 gentle vertical strokes of a perpendicular 400 cm2 paddle. For
the experiments indicated, seeded predators were then added to corrals
(Table 1). Experimental assemblages were allowed to interact in situ for 24
h. Floats suspending the corrals were tethered to the dock. Currents did
not distort corral shape or volume.
Whole fixed samples were sorted at 100x magnification with
epifluorescent microscopy and an FITC filter. Marked larvae were tallied,
fate noted (i.e., in guts or free-living and apparently alive at the time of
fixation), and predator identities recorded. Selected predators were
observed under epifluorescence to check for marked prey, then dissected for
identification and counting of gut contents.
TABLE 1. Summary of observational experiments. Experiments were
conducted at Coos Bay (CB) or Friday Harbor (FH). Seedeq potential
predators are the hydromedusae a) Proboscidactyla, b) Sarsia, and c)
Aglantha, d) the ephyrae of the scyphomedusa Aurelia, e) the chaetognath
Sagitta, f) small postlarval sticklebacks, g) brachyuran zoeae, h) anomuran
zoeae, and i) trochophores of the polynoid polychaete Arctonoe vittata.
Potential predators also include randomly captured animals.
Marked prey (numbers corral-I)
Exp. Site Day/ Sand Bivalve Gastro Seeded potential
Night dollar veligers veligers predators corraI-1
plutei (# in parentheses)
1 CB D 100 100
2 CB D 100
3 FH D 100 100
4 FH D 50 50 a (2), b (1)
5 FH N 100 a:> 50 a (2), e (1)
6 FH N 100 100 a (2)
7 FH N 100 100 a (1), c (1), e (4)
8 FH D 100 100 f(2), d (3)
9 FH D 50 50 d (5), g (5), h (1)
10 FH D 123 123 a (2), d (2), g (3), h (1), i (2)
",-
Potential predators and background plankton were counted for each
corral. Absolute numbers were determined for relatively large potential
predators (> 500 J,1m). Background plankton, including small potential
predators, wild invertebrate larvae, and potential alternative food items for
predators, were counted in 25% sample aliquots. Resulting counts for
background plankton are given as estimated # corral-I. Organisms counted
in 25% aliquots included large diatoms, dinoflagellates, small copepods,
copepod nauplii, barnacle nauplii, wild (unmarked) gastropod and bivalve
veligers, and small polychaete larvae.
Manipulation Experiments
Two corral experiments manipulating natural conditions were
conducted from the dock at Friday Harbor Laboratories, Friday Harbor,
Washington. The first experiment manipulated marked prey density and
the presence of background plankton. The four treatments were 1) near-
natural prey densities in 53-J.Lm-filtered seawater; 2) near-natural prey
densities with background plankton (unfiltered seawater); 3) unnaturally
high prey densities in 53-J.Lm-filtered seawater; 4) unnaturally high prey
densities with background plankton (unfiltered seawater). Experiments
always used pluteus larvae of the sand dollar Dendraster excentricus and
veliger larvae of the oyster Crassostrea gigas as marked prey. The near-
natural prey density used was 0.81 larvae liter-I (Zimmerman, 1972; Miller,
1995). Unnaturally high prey densities were 100 larvae liter-I. Corrals in
treatments with 53-J.Lm-filtered seawater were deployed by lowering them
into the water cod-end-first. Seawater was then screened as it passed
through the cod-end and into the corral, allowing plankton < 53-J.Lm to enter
the corral and excluding larger plankton. After the corral was submerged
to deployment depth, it was suspended from surface floats. Selected
predators were then added to all treatments to determine the effects of prey
density and background plankton on predation rates. Each treatment was
replicated 3 times. All replicates could not be run simultaneously, so one
complete set of the four treatments was run daily for three consecutive
days. All other aspects of this experiment (deployment, collection, and
sorting) were identical to methods described for the observation
experiments.
In the second manipulation experiment, prey densities were held
constant, but the presence of background plankton was manipulated. The
treatments were 1) near-natural prey densities in filtered seawater and 2)
near-natural prey densities with background plankton present (unfiltered
seawater). Near-natural densities were 1 larva liter-l. As with the first
manipulation experiment, corrals in the treatment with 53-JUIl-filtered
seawater were deployed by submerging them cod-end-first. Each treatment
consisted of 3 replicates and the entire experiment took place
simultaneously. All other aspects of this experiment (deplOYment,
collection, and sorting) were identical to methods described for the
observation experiments. Table 2 provides a summary of predators seeded
in both manipulation experiments.
Laboratory Experiments
Laboratory roller table experiments were conducted with the
hydromedusa Proboscidactyla flavicirrata, a predator present in several
100
TABLE 2. Summary of seeded predators in manipulation experiments.
Exp.
1
2
-1Seeded potential predators corral
(# in parentheses)
Proboscidactyla (2), Aurelia ephyrae (2), Muggiaea colony (1),
brachyuran zoeae (3), anomuran zoea (1), Arctonoe
trochophores (6)
Proboscidactyla (2), Aurelia ephyrae (2), small stickleback
(1), brachyuran zoeae (2), anomuran zoea (1), Arctonoe
trochophores (2)
corral experiments. P. fiavicirrata was selected for laboratory investigation
because of the consistent occurrence of large mollusc larvae in the guts of
corral specimens. Experiments were conducted on a roller table (Omori &
Ikeda 1984, Larson & Shanks 1996), which rolled 3-liter cylindrical tanks at
0.75 rpm and prevented plankton from settling. Every 2 hours, tanks were
gently tumbled once and then replaced on the roller table facing the opposite
direction. A single predator was housed in each tank. Though enclosed,
plankton do not suffer oxygen depletion during the experimental time
frame (Larson & Shanks 1996). The roller table was maintained at 12°C in
a constant temperature room with a 14:10 Light:Dark cycle for 24 hours.
Observations of predators an",9 prey in roller tanks revealed that they were
evenly distributed, remained suspended in the water, and exhibited
apparently normal behavior
Hydromedusae were collected in Coos Bay, Oregon at high tide or in
Friday Harbor, Washington and shipped to Oregon for use in roller table
101
experiments. Individual medusae were dipped from the plankton.
Medusae were maintained in filtered seawater and used in experiments
within 2 to 10 days of capture or shipment and exhibited no apparent
behavioral or physiological damage. Veligers (both 90-Jlm and 280-Jlm in
length) of the oyster Crassostrea gigas were used as prey. Oyster larvae
were obtained from Whiskey Creek Oyster Farms, Tillamook, Oregon, and
maintained on a diet of Isochrysis galbana and Rhodomonas sp.. At the
end of each experiment, predators and remaining prey were collected and
fixed with 4% buffered formaldehyde. Counts of prey consumed were made
using a compound microscope with cross-polarized light.
The first P. flavicirrata roller table experiment examined predation
on veligers of the oyster Crassostrea gigas at either of two prey sizes and
each at either of two prey densities. Also, all treatments were conducted in
either filtered seawater or in the presence of whole seawater (with
background plankton). The resulting 8 treatments were each replicated 3
times for a single day's experiment. The entire experiment was repeated
on 2 consecutive days for a total of 6 replicates for each treatment. The
near-natural prey density was 50 1-1 and based upon the highest bivalve
veliger densities in the literature (Carriker, 1951). The second prey density
was unnaturally high (1009' 1-1) and intended to increase encounters and
predation for comparison with predation in the near-natural prey density.
The second P. flavicirrata roller table experiment examined
predation in 3 treatments, all at the near-natural prey density of 501-1• Each
treatment was replicated 6 times. The 3 treatments were 1) d-hinge
102
veligers (90-J..lm in length) as prey, 2) large veligers (280-J..lm in length) as
prey, and 3) both d-hinge and large veligers as prey. Treatment 3 used each
of the two veliger size classes at the full density of 50 I-I.
Results
Marked larvae fluoresced brightly when excited by UV light and
viewed through an FITC filter. Glowing skeletons were visible against the
background plankton from corrals (Figure 2, A and B) and in the guts of
predators (Figure 2, C-F).
Observation Experiments
The high recovery of marked larvae (Table 3) allows reliable
estimates of mortality in the corral. For the 3 marked larval types used in
10 different observational experiments, mean recovery was 99-100% in 15 of
20 cases (1 case = 1 larval type in one experiment). Recovery was 100% in all
replicates in 5 cases. The lowest mean recovery was 96.50% for marked
bivalve veligers in experiment 7. The fate of unrecovered larvae cannot be
determined. All evidence indicates, however, that marked animals are
visible in any condition (i.e., free-living, in predator guts, or in fecal
pellets). Therefore, it is as'Sumed that unrecovered larvae were not more
likely to have been victims of predation than recovered larvae from the same
corrals.
Observations of predation on marked bivalve veligers are
summarized in Table 4. In 4 of 9 experiments using marked bivalve larvae,
103
FIGURE 2. A. Field of view from sample sorting with myriad
phytoplankton and background plankton, viewed under white light. B.
Same view as 'A', observed under epifluorescence (FITC filter) to reveal the
location of a marked pluteus. C. The heterotrophic dinoflagellate Noctiluca
scintillans observed under white light. D. Same view as 'C', observed with
epifluorescence to reveal a phagocytized marked veliger. E. Majid zoea
flattened with slide coverslip and observed under white light. F. Same view
as 'E', observed under epifluorescence. Fluorescent bolus of a crushed
marked pluteus skeleton visible in the zoea's intestine.
104
105
TABLE 3. Mean recovery of marked larvae in observational experiments
(n =4 in all cases).
Mean % recovery (SD)
Exp. Plutei Bivalve vel Gastropod vel
1 100.0 (0.0) 100.0 (0.0)
2 99.5 (0.6)
3 99.5 (1.0) 99.0 (1.4)
4 100.0 (0.0) 98.3 (1.5)
5 98.0 (0.8) 100.0 (0.0) 99.3 (1.2)
6 96.5 (3.4) 97.5 (2.4)
7 99.3 (1.0) 96.5 (4.5)
8 99.7 (0.6) 99.7 (0.6)
9 99.5 (1.0) 100.0 (0.0)
10 99.7 (0.5) 98.9 (1.2)
absolutely no predation on bivalves was observed. This in spite of the fact
that 100% of the larvae were usually recovered. In 9 experiments using
marked pluteus larvae, only a single consumed pluteus larva was observed,
,,-'
preyed upon by P. flavicirrata in experiment 10. No predation on gastropod
veligers was observed.
Table 4 also shows calculations of mortality (day·l) for consistent
observed predators. Instantaneous mortality, M, can be calculated as
follows (Rumrill, 1990):
No is the initial number of prey in a specified water mass. Nt is the final
prey abundance in that same water mass after time t.
Table 4. Predation on marked bivalve veliger larvae from observational
experiments. Instantaneous mortality (M) represents the mean loss to
corral populations in 24 h. In three cases below (4, 6b, and 8) indicated
predators consumed only a single marked bivalve veliger.
Exp. M Estimated loss Predator responsible (total
(day·l) after 28 days number of veligers consumed
in all replicates)
1 0.000 0% No predator (0)
3 -0.070 87%- Noctiluca scintillans (28)
4 -0.005 13% Proboscidactyla flavicirrata (1)
5 0.000 0% No predator (0)
6a -0.035 63% Noctiluca scintillans (14)
(i)
-0.003 7% Spionid metatrochophore (1)
7 0.000 0%"" No predator (0)
8 -0.003 7% Gasterosteus aculeatus (1)
9 0.000 0% No predator (0)
10 -.004 11% Proboscidactyla flavicirrata (2)
106
107
Predators responsible for observed predation on marked bivalves are
indicated in Table 4. In observational experiments 4 and 8, only a single
marked veliger was consumed (i.e., no predation on marked bivalves was
observed in 3 of the 4 replicates). Likewise, in addition to predation by N.
scintillans in experiment 6, a single spionid metatrochophore larva
consumed a marked veliger. Two marked bivalve veligers were observed in
the guts of the hydromedusa Proboscidactyla flavicirrata in experiment 10.
The identity and numbers of background plankton are given in Table 5. In
all, experimental corrals captured dozens of potential predator types
representing a wide variety of planktonic feeding strategies (Greene, 1985).
The following predators were selected for determination of their non-
marked gut contents: all cnidarians, all ctenophores, chaetognaths, fish,
and Arctonoe trochophore larvae. These predators were selected because
they are considered by many to be important predators, their typical prey
sizes encompass the sizes of larvae in corrals, and guts contents could be
viewed or dissected with relative ease. The most common prey items found
in predator guts were copepods and other crustaceans, including the larvae
of copepods and barnacles, and phytoplankton. Predators of copepod and
barnacle nauplii include Sagitta sp., Gasterosteus aculeatus,
Pleurobrachia bachei, and t1le hydromedusae Proboscidactyla flavicirrata,
Sarsia sp., and Phialidium sp.. These same coelenterates were also
observed to contain wild, unmarked bivalve or gastropod veligers.
Polychaete larvae were observed in the guts of P. bachei, P. flavicirrata, and
108
Sarsia sp.. Polychaete numbers were estimated from clusters of undigested
setae.
Manipulation Experiments
Results for the corral experiment which simultaneously
manipulated prey densities and the presence of background plankton are
given for marked plutei (Figure 3A) and marked veligers (Fig1.lre 3B). For
both larval types, predation in filtered seawater at high prey densities was
almost completely eliminated at lower, near-natural prey densities. At the
high prey densities, the inclusion of natural background plankton reduced
predation on plutei by an average of 37% and on bivalves by an average of
23%. When prey were presented at near-natural prey densities and in the
presence of natural background plankton, no predation was observed on
marked larvae. Predators responsible for the predation graphed in Figure
3 included two hydromedusae, a scyphozoan developmental stage, decapod
zoeae, and a polychaete trochophore. These predators are summarized in
Table 6 by prey type and treatment.
Results of the second manipulation experiment were consistent with
the findings of the first manipulation experiment. Predation was generally
low, with the only observed predation on a single marked bivalve veliger in
filtered seawater. When prey (veligers and plutei) were presented under
the most natural conditions (at near-natural prey densities and in the
presence of natural background plankton) no predation was observed on
marked larvae.
109
TABLE 5. Background plankton in observational experiments, including
potential predators randomly captured when corrals were loaded (mean
numbers corral-I). Additional potential predators not shown here include
those seeded in corrals (see experimental setup, Table 1).
Experiment #
Species 1 2 3 4 5 6 7 8 9 10
Protoperdinium 0 0 365 0 0 o 32423 7740 3150 25364
Noctiluca 1531 413 540 489 404 375 0 0 0 0
Coscinodiscus 0 0 0 0 0 o 12938 6420 9923 10025
Tintinnids 193 376 164 0 0 0 0 0 14 0
Copepods (Cal) 653 565 922 737 817 1838 2543 840 1215 1520
Copepods (Harp) 0 101 101 225 72 55 113 16 ~ 21
Copepod nauplii 1806 545 1849 1787 1812 23464 9428 8640 4275 3755
Barnacle nauplii 1439 440 1068 334 268 949 270 43 42 35
Barnacle cyprid 0 13 0 0 0 27 0 8 4 12
Amphipods (G) 0 0 0 0 0 0 0 3 0 0
Amphipods (H) 0 0 0 1 2B 2 9 11 3 5
Cryptoniscid 0 0 0 0 0 0 0 5 1 0
Anomuran zoea 0 0 0 0 2 0 2 0 4 1
-.,-
Brachyuran zoea 0 0 0 0 0 2 1 0 5 3
Megalopa 0 0 0 0 0 0 0 0 0 0
Cladoceran 0 0 11 8 1 0 0 0 0 0
TABLE 5. Continued.
110
Experiment #
Species 1 2 3 4 5 6 7 8 9 10
Ostracoda 0 0 0 0 0 0 1 3 0 0
Cumacea 0 0 0 0 0 0 5 0 0 0
Euphausid zoea 0 1 0 0 1 1 0 0 0 0
Salt water mite 0 0 2 0 0 0 0 0 0 0
Obelia 21 0 22 35 3 0 2 0 3 1
Phialidium 0 1 0 1 0 1 3 0 1 0
Aglantha 0 0 0 0 0 0 0 1 0
Leptomedusa 0 42 0 8 1 0 2 0 19 0
Rathkea 0 1 0 0 0 0 0 0 0 0
Pleurobrachia 1 2 1 0 0 0 0 0 1 1
Cydippid larvae 45 0 0 0 0 0 0 0 0 0
Autolytus 0 0 0 0 8 4 5 3 4 1
Spionids 0 0 0 0 0 0 228 35 ~ 11
Metatrochophore 792 259 '\<- 415 24.0 167 662 70 13 7 3
Nectochaeta 0 0 0 0 0 0 7 0 0 0
Mitraria 8 0 0 0 0 0 0 0 0 0
Magelona 42 ro 16 0 0 0 0 0 0 0
111
TABLE 5. Continued.
Experiment #
Species 1 2 3 4 5 6 7 8 9 10
Trochophores 34 42 8 12 24 'Zl 0 0 0 0
Cyphonautes 0 3 6 22 15 17 5 8 4 0
Pilidia 16 0 0 0 0 0 0 0 0 0
Doliolaria 0 3 0 0 0 0 0 0 0 0
Ophioplutei 0 0 14 0 0 0 0 0 0 0
Echinoplutei 241 48 5 0 23 0 0 0 0 0
Little urchins 7 0 0 0 0 0 0 0 0 0
Veligers (Biv) 404 83 112 0 0 192 96 115 126 48
Veligers (Gast) 13 ~ 9 00 22 00 9 19 8 15
Egg cases 0 0 0 0 4 3 0 0 0 0
Embryos 216 148 53 0 0 95 0 0 0 0
Eggs 92 331 0 0 48 4 0 0 0 0
Chaetognaths ID ~v 22 9 67 71 62 32 15 13
Larvaceans 234 9 89 114 102 55 16 8 31 ID
Larval fish 0 0 0 0 0 0 1 0 0 0
112
50
A.
D45 FSW
~ 40
•~- 35
BG
.1"'4 .-4
-.
=-~ =30~ ~o ~
e 8 25
m=*t:: 20
==~ ~ 15l! e~- 1
5
0 0B.
30
~~ 25.1"'4 •
--==~ ~~ ~
o 0e Q
~=*t:: 1~ =~=:.= ~ 1~!
O-f-_~....~O__--f----L__
Prey density of 1 Prey density of 100
larva liter- l larvae liter-l
FIGURE 3. The effects of prey density and background plankton on larval
mortality for A. marked plutei and B. marked veligers. Clear bars are 53-
Jlm-filtered seawater treatments (FSW). Black bars are background
plankton treatments (BG). Cases of zero mean and variance in mortality
are indicated by a '0'. Error bars are 95% Confidence Intervals.
113
TABLE 6. Manipulation experiment: predators on marked pluteus (P) and
marked veliger (V) larvae by treatment. Experiments were conducted at
each prey density either in 53-J..lm-filtered seawater (fsw) or with
background plankton present (BG). (Total number of prey consumed in all
3 replicates given in parentheses).
Proboscidactyla V{l) 0 P(83)N{57) P(13)N{4)
Phialidium 0 0 0 V(4)
Aurelia ephyra 0 0 P(4)N(l) P(l)
Brachyuran zoea P{l) 0 P(17) P{l)
Anomuran zoea 0 0 P(2) P{l)
Arctonoe 0 0 V(l6) 0
BG
High prey density
fswBG
Low prey density
fsw
114
Laboratory Experiments
The hydromedusa Proboscidactyla flavicirrata consumed small and
large bivalve veligers under all conditions (Figure 4). Small (d-hinge)
veligers were consumed most frequently when presented at unnaturally
high prey densities in filtered seawater (mean = 60.5 larvae gut-I). Large
veligers, however, were not consumed significantly more when presented
at high prey densities and in filtered seawater than in other treatments.
Presenting prey in the presence of background plankton at the high prey
density dramatically reduced predation on d-hinge veligers from 60.5 gut- l
to 15.5 gut-I. When offered a choice of d-hinge and large veligers, each at
the same prey densities, P. flavicirrata reduced predation on d-hinge
veligers from 10.0 gut-I to 1.5 gut- l in favor of persistent predation on large
bivalve larvae (Figure 5).
Discussion
Predation on marked plutei and veligers in these experimental
assemblages was low in most cases. This may be due to the exclusion of
important natural predators, though dozens of potential predators
representing diverse foraging strategies (Greene, 1985) were captured or
-..-
seeded in corrals. Thus, we seek alternative explanations to account for the
lack of predation. Infrequent encounters at near-natural prey densities
may reduce predation by predators which might otherwise consume larvae.
Also, naturally occurring background plankton may somehow interfere
115
.... D~ 80 FSW&
ell 70
•
BG~
Q)
bll 60;.=
~ 50
Q)
:;. 40
-~ 30
."",
,.Q
=
20
~
~ 10
O-+-~-
Small Large
Prey density of 50
larvae liter-!
Small Large
Prey density of 1000
larvae liter-!
FIGURE 4. Mean number of bivalve veligers consumed by the
hydromedusa Proboscidactyla flavicirrata in laboratory experiments. The
treatments are: 53-Jlm-filtemd seawater (FSW) with small bivalve veligers
as prey; natural background plankton (BG) with small bivalve veligers as
prey; 53-Jlm-filtered seawater (FSW) with large bivalve veligers as prey;
natural background plankton (BG) with large bivalve veligers as prey.
Error bars are 95% Confidence Intervals.
16
..... 14~
Sn
l1.l 12~
~ 10..........
~
l4-o 80
=1:1::
~ 6Cd
C1>
::;s 4
2
0
Small
Veligers
Alone
Large
Veligers
Alone
Small and Large
Veligers
(Mixed Prey)
116
FIGURE 5. Laboratory predation by Proboscidactyla flavicirrata on small,
large, and mixed (large and small) bivalve veligers. Each prey size was
always presented at the neM-natural high density of 50 larvae liter-I. Error
bars are 95% Confidence Intervals.
117
with predation in the following ways: predators may spend time handling
or consuming background plankton, reducing encounters with larvae;
predators may become satiated after consuming background plankton;
background plankton may obscure larvae from detection or capture.
Whatever the explanation, potentially low planktonic predation rates
challenge what has become a paradigm in marine invertebrate life history
theory-that predation on meroplanktonic invertebrate larvae is high.
We recognize that our experiments were only for 24 h, while many
invertebrate larvae are in the plankton for weeks to months. Several
experiments showed no predation in any replicate in 24 h. If we assume
that, given another 24 h, a single predation event would have occurred,
then instantaneous mortality would be -0.00125 day·l. After 28 days in the
plankton, this maximum estimate of mortality in these corrals results in a
population loss of only 3.4%. Thus, of a single female's 1.0 x 105 fertilized
offspring, 9.7 x 104 would survive predation and be available for
recruitment.
The manipulation experiments provide clues to why predation was
infrequent in the observational experiments. Many captured and seeded
predators did prey on marked veligers and plutei when prey were presented
at unnaturally high densiti~s or in the absence of natural background
plankton. Natural treatments mimicking the observational experiments,
however, produced results consistent with observational experiments and
reduced or eliminated predation. Johnson and Shanks (1997) and Johnson
and Brink (1998) made similar observations in laboratory studies of
118
planktonic predation on echinoid embryos, plutei, barnacle nauplii, and
bivalve veligers. Predation on echinoid and barnacle larvae by
leptomedusae and anomuran zoeae, prevalent when prey were presented at
unnaturally high densities or in filtered seawater, was reduced or
eliminated under the most natural conditions (Johnson and Shanks, 1997).
Likewise, the most natural laboratory conditions reduced or eliminated
predation by 4 types of larval polychaetes on bivalve veligers (Johnson and
Brink, 1998). An analogous study investigated the effects of nongrazeable
particles, similes for natural background plankton, on prey capture by a
tintinnid, a rotifer, a gastropod veliger, and young copepods and found that
background could affect feeding rates (Hansen et al., 1991).
The simplest explanation for reduced predation at near-natural prey
densities is that the predators and prey do not encounter one another at
such low densities. Alternatives to low encounters include background
plankton interference and changes in predator behavior at near-natural
larval dentsities. We used the Gerritsen and Strickler (1977) model to
estimate predator-prey encounters in corral experiments. This encounter
model uses predator encounter radius R, prey density Nh, and predator
and prey swimming speeds, v and u, respectively, to determine the number
of encounters Zp of a single 'Predator with its prey for v ~ u:
119
This formula was used to estimate for each experiment the mean total
encounters of marked larvae with potential predators in a corral. Potential
predators are those which were seeded in corrals plus those background
predators which might prey on larvae (e.g., relatively large animals).
Potential predators along with estimates of their body radiuses and
swimming speeds are presented in Table 7. For all predators, body radius
was used as a minimum encounter radius. Some predators may sense
prey at a distance using remote visual, chemical, or vibratory sensory
mechanisms (Horridge and Boulton, 1967; Giguere and Northcote, 1987;
Giske et al., 1994). Determination of remote encounter radiuses is
complicated (Gerritsen and Strickler, 1977) and for simplicity in these
estimates we used the minimum estimate of encounter radius-the radius
of the predator's body. Estimated encounters are given for marked plutei
(Table 8) and marked bivalve and gastropod veligers (Table 9) for each
experiment. Encounter estimates given are the mean number of larvae
expected to be consumed (if successful capture rate is 100%) corraI-1 for each
of the potential predators.
Corral totals are tallied for all potential predators. Because little is
known about natural predator-prey relationships in the plankton, some
predators may not have the'l1bility to prey on these larval types.
Unfortunately, not enough information is available to confidently exclude
many questionable potential predators. We therefore offer two sets of
estimates to evaluate potential predation by all possible predators as well as
by a subset of more likely predators. Our subset of predators, used to
120
TABLE 7. Predator encounter radiuses (R) and predator and prey
swimming speeds (v and u, respectively) used in calculating encounter
estimates for Tables 9 and 10. 'R' is a minimum estimate of encounter
radius-predator body radius.
Predators R (em) v (cm/s) Source for lV'
Calanoid copepods 0.02 1.2 Strickler, pers. comm.
Harpacticoid copepods 0.02 0.6 personalobs.
Gammarid amphipods 0.04 1.0 personal obs.
Hyperiid amphipods 0.04 1.5 personalobs.
Cryptoniscid 0.01 0.6 personalobs.
Anomuran zoeae 0.03 0.9 Knudsen, 1960; Latz
and Forward, 1977;
Cronin and Forward,
1980; Forward and
Cronin, 1980; Sulkin,
1973;Sulkin, 1975
Brachyuran zoea 0.03 0.9 Knudsen, 1960; Latz
and Forward, 1977;
Cronin and Forward,
1980; Forward and
Cronin, 1980; Sulkin,
1973;Sulkin, 1975
Cumacea 0.06 2.0 personal obs.
Euphausid zoea 0.04 1.5 personal obs.
Obelia .0.02 0.5 approx. sinking rate
Phialidium 0.3 0.5 approx. sinking rate
Aglantha 0.2 2.0 personal obs.
Leptomedusa 0.03 0.5 approx. sinking rate
TABLE 7. Continued.
Predators R (em) v (cm/s) Source for 'v'
Rathkea 0.03 0.5 approx. sinking rate
Pleurobrachia 0.4 0.5 approx. sinking rate
Cydippid larvae 0.01 0.6 personal obs.
Proboscidactyla 0.3 0.5 approx. sinking rate
Sarsia 0.1 0.5 approx. sinking rate
Aurelia ephyra 0.06 0.2 personal obs.
Autolytus 0.02 1.9 personal obs.
Spionid metatrochophores 0.02 0.1 Konstantinova, 1969
Misc. metatrochophores 0.04 0.1 Konstantinova, 1969
Nectochaeta 0.03 0.1 KDnstantinova, 1969
Magelona 0.02 0.3 personal obs.
Misc. trochophores 0.04 0.2 Konstantinova, 1969
Arctonoe trochophore 0.02 0.3 Pernet, pers. comm.
Chaetognaths 0.1 0.3 personal obs.
Larval fish 0.2 1.2 personal obs.
Prey u Source for 'u'
Pluteus larvae N/A 0.015 personal obs.
Bivalve veliger larvae N/A 0.03 Hidu & Haskin, 1978
Gastropod veliger larvae N/A 0.09 Konstantinova, 1966
121
122
estimate conservative encounters, included only predators observed in the
laboratory to consume the larvae.
Two observed predators, the heterotrophic dinoflagellate N.
scintillans (experiments 3 and 6) and juveniles of the threespine stickleback
Gasterosteus aculeatus (experiment 8), were potentially of great
importance to their prey populations. These predators each have a life
history unique from that of our other potential predators. In the case ofN.
scintillans, predation on veligers is yet another example of how protists can
upset the paradigms of traditional food webs (Capriulo et al., 1991; Jeong,
1994; Glasgow et al., 1995). The threespine stickleback G. aculeatus is an
effective predator on nauplii, but, even with 1 animal, was probably over-
represented in corrals. Noctiluca scintillans predation in experiments 3
and 6 and G. aculeatus predation in experiment 8 are not addressed in
Tables 9 and 10, but are examined in depth later in the discussion.
Mean estimated encounters for the observational experiments, where
prey densities were intended to reflect natural densities, ranged from near
o to as high as 269 (encounters of prey with large calanoid copepods in
experiment 7) for only one predator type. The highest overall encounter
rate was in experiment 7 where the calculations suggested 354 encounters.
The fewest overall estimate<l encounters was 24.5 in experiment 5. Actual
predation due to these predators was, however, completely absent in
experiments 1, 2, 5, 7, and 9, and nearly absent in experiments 3, 4, 6, 8,
and 10. The sum of observed predation on plutei for all experiments was 0.3
corraP. The sum of conservative estimates (i.e., estimates of encounters
123
Estimated pluteus encounters with potential predators (corral-I)
TABLE 8. Estimates of marked pluteus encounters with predators in
observational experiments. These numbers are corral- l estimates (the
mean of estimates from all replicates for a single experiment). Seeded
predator names are underlined. Underlined numbers are the estimates of
predators confirmed to prey on plutei under some conditions (e.g., in the
laboratory, in filtered seawater, or with unnaturally high prey densities).
Underlined numbers are used to calculate more conservative corraI-1
encounter estimates based on the predators confirmed ability to prey on
plutei. Known encounters of predators with marked plutei in corrals (i.e.,
observed predation) is given for comparison with mathematical estimates.
109876
Experiment #
2351Potential Predator
Copepods (Cal) 00 00 00 En 195 269 89 64 198
Copepods (Harp) 0 1 1 1 1 2 0 0 0
Amphipods (G) 0 0 0 0 0 0 0 0 0
Amphipods (H) 0 .0 0 4 0 1 1 0 1
Anomuran zoeae Q Q Q Q Q Q Q Q Q
Brachyuran zoeae Q Q Q Q Q Q Q 1 1
Euphausiid zoeae 0 1 0 1 0 0 0 0 0
Obelia 1 0 1 0 0 0 0 0 0
Phialidium o yIO 0 0 12 ro 0 4 0
Aglantha 0 0 0 0 0 ID 0 10 0
Leptomedusa 0 12 0 0 0 1 0 3 0
Rathkea 0 0 0 0 0 0 0 0 0
TABLE 8. Continued.
Estimated pluteus encounters with potential predators (corral-I)
Experiment #
Potential Predator 1 2 3 5 6 7 8 9 10
Proboscidactyla .Q .Q .Q ID ID 10 .Q .Q 24
Barsia 0 0 0 0 0 0 0 0 0
Aurelia ephyrae .Q .Q .Q .Q .Q .Q 1 .Q .Q
Pleurobrachia 18 .9 13 .Q .Q .Q .Q 1 22
Cydippid larvae 1 0 0 0 0 0 0 0 0
Autolytus 0 0 0 1 1 1 1 0 0
Spionids 0 0 0 0 0 2 0 0 0
Misc. metatrochs 31 10 16 7 ~ 3 1 0 0
Nectochaeta larvae 0 0 0 0 0 0 0 0 0
Magelona 1 1 0 0 0 0 0 0 0
Misc. trochophores 2 3 1 2 2 0 0 0 0
Arctonoe 0 0 0 0 0 0 0 0 0
Chaetognaths 0 0 0 0 1 0 0 0 0
'\t'./
Larval fish 0 0 0 0 0 5 0 0 0
124
Total estimated 122 107 130 122 257 354 00 00 247
Observed predation 000 0 o o o o o
125
Estimated encounters corral-1 of veliger larvae with potential predators
Table 9. Estimates of marked veliger encounters with predators in
observational experiments. These numbers are corral"l estimates (the
mean of estimates from all replicates for a single experiment). Seeded
predator names are underlined. Underlined numbers are the estimates of
predators confirmed to prey on the veligers under some conditions (e.g., in
the laboratory, in filtered seawater, or with unnaturally high prey
densities). Underlined numbers are used to calculate a more conservative
corral-1 estimate of contact with potential predators. Known encounters of
marked veligers with potential predators (i.e., observed predation) are given
for easy comparison with mathematical estimates. Predation on bivalve
veligers by N. scintillans in experiments 3 and 6 is later in the discussion.
Gastropod
Veligers
Experiment #
Bivalve
Veligers
Potential predator 1 3 4 5 6 7 8 9 10 4 5
39 43Copepods(Cal)
Copepods (Harp)
Amphipods (G)
Amphipods (H)
Anomuran zoeae
Brachyuran zoeae
Cumaceans
Euphausiid zoeae
Obelia
Phialidium
00 00 39 17 195 200 ffi 64 198
012012000
000000000
000001101
000000000
000000011
",-
o 0 0 0 0 10 0 0 0
000000000
111000000
!!!! Q !!12 OO!! 1!!
2
o
o
o
o
o
o
1
5
1
o
2
o
o
o
o
o
o
126
TABLE 9. Continued.
Estimated encounters corral"l of veliger larvae with potential predators
Bivalve Gastropod
Veligers Veligers
Experiment #
Potential predator 1 3 4 5 6 7 8 9 10 4 5
Aglantha 0 0 0 0 0 ID 0 10 0 0 0
Leptomedusa 0 0 1 0 0 0 0 3 0 1 0
Rathkea 0 0 0 0 0 0 0 0 0 0 0
Proboscidactyla .Q .Q 10 i ID 10 .Q .Q 24 10 10
Sarsia 0 0 1 0 0 0 0 0 0 1 0
Aurelia ephyrae .Q .Q .Q .Q .Q .Q 1 .Q .Q 0 0
Pleurobrachia 18 13 .Q .Q .Q .Q Q 1 22 0 0
Cydippid larvae 1 0 0 0 0 0 0 0 0 0 0
Autolytus 0 0 0 0 1 1 0 0 0 0 1
Spionids .Q .Q .Q .Q .Q .2 .Q .Q .Q 0 0
Misc. metatrochs 32 17 5 1 ~ 3 1 0 0 6 4
v
Nectochaets 0 0 0 0 0 0 0 0 0 0 0
Magelona 1 .Q .Q .Q .Q .Q .Q .Q .Q 0 0
Misc. trochs 2 1 0 0 2 0 0 0 0 0 1
Arctonoe .Q .Q .Q .Q .Q .Q .Q .Q .Q 0 0
127
TABLE 9. Continued.
Estimated encounters corral-1 of veliger larvae with potential predators
Bivalve Gastropod
Veligers Veligers
Experiment #
Potential predator 1 3 4 5 6 7 8 9 10 4 5
Chaetognaths 0 0 0 0 1 0 0 0 0 0 0
Larval fish 0 0 0 0 0 5 0 0 0 0 0
Total estimated 123 131 63 25 258 354 00 00 247 64 62
Observed predation 0 0 0 0 0 0 0 0 1 o o
128
with confirmed laboratory predators) of pluteus-predator encounters was
146 encounters corral-I. Assuming estimates are correct, this indicates an
average capture success rate of 0.21%. Similar comparisons for encounter
estimates with bivalve veligers show that, for all predators excepting N.
scintillans, average capture success rate is 0.54%.
Assuming swimming speeds and encounter radiuses used in
calculations are accurate, then low predation may be due to low capture
success rates, predator preferences, predator inability to consume larvae,
or the effects of background plankton. For example, the hydromedusa
Proboscidactyla fiavicirrata is known, based upon gut content data, to be a
predator of wild bivalve veligers (C. Mills, pers. communication).
Estimates suggest that this medusa had many opportunities to consume
(i.e., encounters with) marked veligers in corrals. In spite of the estimates,
however, only 3 marked bivalve veligers were consumed (l in experiment
#4 and 2 in experiment #10). P. fiavicirrata's capture success rate is
unknown. Consequently, it is possible that all estimated contacts were
made and the predator failed to retain the prey. An alternative explanation
is that P. fiavicirrata prefers and somehow selects other prey. Wild bivalve
veligers consumed by P. fiavicirrata in corrals were much larger (250-350-
JlIIl) than the marked d-hinge oyster veligers (90-Jlm) added to corrals. Our
laboratory experiments showed, P. fiavicirrata feeds selectively on 280-Jlm
veligers over 90-JlIIl d-hinge veligers, supporting the idea that prey selection
may be partially responsible for low predation on the added small marked
veligers.
129
Some animals we have labeled as "potential predators" may actually
lack the ability to consume the larvae in question. For example, it is
unknown whether the abundant large calanoid copepods in corrals are
omnivorous or strictly herbivorous. It is therefore possible that these
potential predators, often responsible for over half of estimated encounters,
would not prey on larvae under any circumstances.
Finally, It has been hypothesized that background plankton can
reduce encounters between predators and larvae, obscure larvae from
detection or capture, or serve as substitute food, occupying or satiating the
predator (Johnson and Shanks, 1997). Indeed, the lack of predation in
background plankton treatments (manipulation experiments) may reflect
background plankton's influence. If so, planktonic encounter models
should incorporate the potential effects of background plankton on
encounters. Assuming that high encounter estimates are accurate, that
all potential predators consume larvae given the opportunity, and that prey
capture success is high, background plankton plays a significant role in
reducing predation on invertebrate larvae.
Some predators did, however, consume wild unmarked larvae.
Moderate predation was observed on bivalve veligers and more substantial
predation was observed on tlie nauplius larvae of barnacles and copepods.
Encounter estimates, analogous to those calculated for marked larvae,
predict that predators had ample opportunity to consume wild larvae.
Numbers of wild nauplii far exceeded the number of marked larvae in
corrals. For example, the corral average of 23,464 copepod nauplii
130
(experiment 6) is more than 2 orders of magnitude greater than the marked
pluteus and veliger densities (100 corral-I). This probably reflects a natural
disparity in the larval abundance; pluteus and veliger densities were never
observed to be as high as the natural crustacean larva densities captured in
the corrals. Consequently, simple encounters, but with a low capture
success rate, could explain the higher predation on nauplius larvae by
many predators. For example, a failed capture attempt on a pluteus larva
may be that predator's only pluteus encounter. On the other hand, a failed
attempt to capture a nauplius is sure to be only one of many opportunities.
In the experiment with the high nauplius densities quoted above, an
average of 4 copepod nauplii corral-I were consumed by Phialidium
medusae. This apparently high predation, however, represents an
instantaneous mortality of only -0.00017 day-I because of the enormous
population size. If this mortality were constant for 90 days in the plankton,
only 1.5% of the nauplius larva population would be lost.
Extremely high prey abundance may influence predator behavior
and the evolution of foraging, providing alternative explanations for the
disparity between predation on plutei and veligers vs. copepod and barnacle
nauplii. Predators may seek out prey and become adept at prey capture
when prey are abundant (Mook et al., 1960; Tinbergen, 1960; Gibb, 1962;
Murton, 1971). Predators may evolve effective foraging and capture
strategies specifically targeted towards prey which are abundant and
consistently available in evolutionary time. Nauplii and other planktonic
crustaceans can reach densities of 100,000's or more m-3 (e.g., Zimmerman,
131
1972). In addition to high abundance relative to echinoid and mollusc
larvae, nauplii and other crustaceans swim with jerky movements,
attracting the attention of visual and vibration-sensing predators (Horridge
and Boulton, 1967; Feigenbaum and Reeve, 1977; Bailey and Yen, 1983; Yen,
1987; Yen and Nicoll, 1990; De Mott and Watson, 1991). Nauplii may have
more predators and stronger predator-prey relationships than the other
larval types investigated.
To help explore predation on wild larvae in corrals, we estimate
encounters of several wild larval types and their predators (Table 10).
Observed predation is also presented in Table 10 for comparison with
encounter estimates. Once again, estimates of encounters are far higher
than observations of predation. The same explanations offered for this
observation with marked larvae also apply here. Support for an encounter-
based explanation includes the relative increase in predation on nauplii
when compared to the far less abundant veligers and plutei.
The stickleback Gasterosteus aculeatus, starved for 3 days and
seeded as a predator in observational experiment 8, had a diet comprised
almost entirely of crustacean larvae. Following the corral experiment, the
guts of G. aculeatus contained an average of 58 copepod nauplii (sd=19.8)
and 22 barnacle nauplii (sd~6.3) each. Total nauplius abundance in the
corrals was an average of 8640 and 43 for copepod and barnacle nauplii,
respectively. Observed predation by G. aculeatus produced instantaneous
mortality of -.0067 day·! for copepod nauplii and -0.51 day·! for barnacle
nauplii. Therefore, mortality due to three-spined sticklebacks would result
132
Table 10. Encounter estimates of specified predators with wild invertebrate
larvae in corral assemblages over 24 h. Predators, prey, and experiments
scrutinized are based upon observations of predators consistently
consuming one or more of these larval types. Observed predation is
presented immediately below encounter estimates for comparison. 'X'
indicates the predator was not present in that experiment (or, in some
cases, that the predator was only present in one or two replicates).
Experiment number
Predators on copepod nauplii 3 4 5 6 7 10
Estimates 285 X X X X 771
Pleurobrachia Mean observed 0 X X X X 4
Capture rate (%) 0 X X X X 0
Estimates X 413 418 5419 1089 Em
Proboscidactyla Mean observed X 0 0 2 1 0
Capture rate (%) X 0 0 0 0 0
Estimates X ~ X 3387 3266 X
Phialidium Mean observed X 1 X 4 1 X
Capture rate (%) X 1 X 0 0 X
Predators on barnacle nauplii
",-
Estimates 164 X X X X 7
Pleurobrachia Mean observed 2 X X X X 0
Capture rate (%) 1 X X X X 0
TABLE 10. Continued
Experiment number
Predators on barnacle nauplii 3 4 5 6 7 10
Estimates X 77 82 219 31 8
Proboscidactyla Mean observed X 0 0 0 0 0
Capture rate (%) X 0 0 0 0 0
Estimates X 39 X 137 94 X
Phialidium Mean observed X 0 X 0 0 X
Capture rate (%) X 0 X 0 0 X
Predators on gastropod veligers
Estimates 1 X X X X 3
Pleurobrachia Mean observed 0 X X X X 0
Capture rate (%) 0 X X X X 0
Estimates X 10 4 6 1 3
Proboscidactyla Mean observed X 0 0 0 0 0
Capture x:,~te (%) X 1 0 2 0 3
Estimates X 5 X 4 3 X
Phialidium Mean observed X 0 X 0 0 X
Capture rate (%) X 4 X 0 0 X
133
TABLE 10. Continued
Experiment number
Predators on bivalve veligers 3 4 5 6 7 10
Estimates 15 X X X X 9
Pleurobrachia Mean observed 0 X X X X 0
Capture rate (%) 0 X X X X 0
Estimates X 0 0 38 10 10
Proboscidactyla Mean observed X 1 1 1 0 1
Capture rate (%) X 2 1 8
Estimates X 0 X 2A 29 X
Phialidium Mean observed X 1 X 0 1 X
Capture rate (%) X X 0 4 X
134
135
in a 17% loss of copepod nauplii in the corral after 28 days. This is
substantial predation, but does not compare to the impact of G. aculeatus on
the corrals' barnacle nauplius populations-100% consumed in 8 to 9 days!
Barnacle nauplii were strongly preferred over copepod nauplii based upon a
comparison of gut and corral nauplius ratios (chi-square goodness-of-fit
test, a. = 0.0001). Of course, major predators like G. aculeatus are not
abundant compared to smaller potential predators we have examined.
Also, though these G. aculeatus were small, they still swim quickly.
Outside of an enclosure one might not expect fish such a G. aculeatus to be
a threat due to short residence times.
Other predators have developed search strategies that favor
crustacean prey. Chaetognaths and some copepods hunt by sensing
vibrations in their prey (Horridge and Boulton, 1967; Feigenbaum and
Reeve, 1977; Bailey and Yen, 1983; Yen, 1987; Yen and Nicoll, 1990; De Mott
and Watson, 1991). Animals -hunting in this manner may detect nauplius
prey at relatively great distances, and yet ignore near-by ciliated swimmers
such as veliger and pluteus larvae. Predation on large wild veligers and
nauplii supports the idea that predation can be specific to larval type and
vary with larval stage (Rumrill et al., 1985; Pennington et al., 1986).
Marine invertebrate life history evolution is influenced by planktonic
mortality rates. Evidence of high planktonic survivorship weakens
arguments which explain phylogenetic patterns of larval feeding in terms
of high planktonic mortality. If it can be shown that circumstances
commonly exist where planktonic larvae suffer little or no predation, then
136
other sources of mortality may be necessary to continue supporting
evolutionary arguments for short planktonic period driven by high
mortality. Alternative explanations for patterns of evolution in feeding
mode are bolstered by observations of low predation. Low mortality in the
plankton lends support to alternative explanations, such as the "use-it-or-
lose-it" hypothesis.
Other discussions of life history theory are indirectly affected by the
possibility that the plankton can be a low risk environment. One unresolved
argument explaining the unique evolution of long planktonic larval periods
is the possibility that dispersal reduces the risk of extinction (Strathmann,
1974; Strathmann, 1990). Long-lived planktonic larvae can be carried 10's
or 100's of kilometers by ocean currents. In general, long-lived planktonic
larvae undoubtedly experience greater dispersal than short-lived species
because long-lived larvae are carried greater distances by ocean currents.
Possible benefits of wider dispersal include more genetic variation within
populations, less inbreeding, less need for simultaneous hermaphroditism,
wider geographic range, and fewer extinctions (Strathmann, 1990). These
potential long-term population benefits arising from long-distance
dispersal might be overwhelmed and lost if predation rates in the plankton
are high. If mortality in the"plankton is low, however, then populations
may be free to experience unfettered selection even for subtle benefits of
remaining in the plankton for dispersal.
In this discussion of our results, we have until now discussed the
majority of experiments which consistently yielded low or nonexistent
137
predation. Two experiments, however, revealed substantial predation on
marked bivalve veligers by the heterotrophic dinoflagellate Noctiluca
scintillans. This predation represented a substantial threat to the veliger
population. N. scintillans is notorious for consuming a diverse range of
prey (Enomoto, 1956; Prasad, 1958; Hattori, 1962; Kimor, 1979; Kirchner et
al., 1996), including metazoans with longer body lengths than the predatory
cell's diameter (Prasad, 1958). It is still some surprise, however, that this
slow-moving protist can capture and engulf oyster larvae. Mean
instantaneous mortality due to N. scintillans was -0.07 and -0.04 day-l for
experiments 3 and 6, respectively. These instantaneous mortality rates
extrapolated over a 28 day planktonic period, a reasonable developmental
time for many bivalves, would produce total population losses of 87% and
68%, respectively. We estimate encounters between N. scintillans and
marked bivalve veligers using another Gerritsen and Strickler equation
(Gerritsen and Strickler, 1977) prepared specifically for cases where
predator swimming speed is less than that of its prey. For the swimming
speed 'v' of N. scintillans, we use its ascension rate. Ki~rboe and Titelman
(1998) contend that ascension (-1 m h-1 or 0.028 cm S-I) is the mechanism of
active foraging for N. scintillans. They also state that N. scintillans'
collects prey with a large mass of sticky mucus attached to the tentacle.
Though N. scintillans feeds primarily on immobile prey such as diatoms
(Enomoto, 1956; Prasad, 1958; Ki~rboe and Titelman, 1998), there are also
observations of predation by this dinoflagellate on metazoans (Enomoto,
1956; Prasad, 1958; Hattori, 1962; Kimor, 1979). We observed (Figure 6) N.
138
scintillans capturing bivalve veligers by 'raptorial' feeding with its modified
tentacle (Omori and Hamner, 1982). The mucus at the tip of the tentacle
(Kil/Srboe and Titelman, 1998) not only assists in capture, but may increase
the encounter radius. The combined length of the tentacle and mucus
string from the cell in Figure 6 is 170-f.lm. As cells ascend and forage, their
orientation appears to be random when a long mucus strand is not present
(Kil/Srboe and Titelman, 1998). We, therefore, consider the ascending cell to
be oriented for capturing encountered prey only 50% of the time and divide
estimated encounters in half to compensate. Prey density was the known
density of marked veligers, 0.811-1. The predatory dinoflagellate was at
relatively low densities in these experiments, 4.5 and 3.1 cells liter-1for
experiments 3 and 6, respectively. The resulting encounter estimates
between N. scintillans and marked bivalve veligers were 0.69 and 0.47
corra!"1 for experiments 3 and 6, respectively. Assuming 100% capture
success, extremely unlikely for this predator-prey combination, an
encounter radius of 0.039 cm is required to accurately predict predation.
Mucus strands several mm long and -100-f.lID wide have been observed, but
result in a downward orientation (i.e., the tentacle points downward) of
ascending cells (Kil/Srboe and Titelman, 1998). This fishing strategy,
analogous to the dragging or tentacles by many coelenterates, may require a
different model for estimating encounters.
Another possibility to consider is that N. scintillans and bivalve
veligers are concentrated together somewhere within the corral. It is
possible that N. scintillans aggregates at the surface of the water column
139
FIGURE 6. Capture of a bivalve veliger by the heterotrophic dinoflagellate
Noctiluca scintillans. A. The waving tentacle of N. scintillans first strikes
the body of a newly encountered bivalve veliger. The veliger is indicated by a
white arrow and the black arrow points to the base of the tentacle. B. After
first contact the bivalve veliger changes direction and swims to escape a
possible predator. C. Attached to the tentacle by an invisible strand of
mucus, the swimming veliger actually drags the large cell through the
water. D. Eventually the prey is reeled in and the tentacle can then
manipulate the bivalve. The tentacle repeatedly strokes, drawing the
veliger through a food groove, where it will eventually become wedged near
the cytostome. E. The veligers is phagocytized at the cytostome. Cross-
polarized light causes the calcium carbonate skeleton of the ingested veliger
to stand out against the faded cell.
.....
J
I
I
\
t.
J
140
141
due to the positive buoyancy of cells in seawater (Kesseler, 1966) and the
dampened turbulence within corrals. Bivalve veligers concentrate at the
surface in stagnant laboratory culture containers (personal observation).
Sampling of different depths within the corrals to locate marked larvae
indicated that oyster veligers do not aggregate at, and in fact may avoid, the
surface. Whatever the mechanism of encounter, N. scintillans can
consume bivalve larvae (Figure 6).
Large wild bivalve veligers (250 to 350-Jlm) were also present in
experiments 3 and 6 at an average of 112 and 192 corral-I, respectively, and
slightly exceeded the numbers of marked oyster veligers. These bivalve
larvae, however, were never observed inside of N. scintillans in spite of a
deliberate examination of N. scintillans cell contents. It was not practical,
however, to examine every N. scintillans cell. Therefore, predation rates
rivaling those on marked bivalve veligers might apply to the wild veligers in
corrals. N. scintillans only preying on small veligers, if true, might be
explained by attributes unique to wild veligers. Bivalve larvae 250-JlID or
larger may be too large for ingestion by N. scintillans, though the cells in
corrals ranged in size from 500 to 1000-JlID. Another attribute related to
increased size is greater swimming speed, which may give veligers the
ability to pull free from the"'dinoflagellate's entrapping mucus.
A predator which may also be responsible for natural mortality of
bivalve veligers is the hydromedusa Proboscidactyla flavicirrata, a
consistent corral predator on marked veligers, which was second only to N.
scintillans in its potential impact on veliger populations. This agrees with
142
unpublished data on the gut contents of 10 species of wild-caught
hydromedusae in Friday Harbor, Washington, collected by Claudia Mills.
P. flavicirrata was the only species that consumed mollusc larvae. Bivalve
and gastropod veligers made up 65-80% of P. flavicirrata gut contents by
number in the late Spring and 65% in the Autumn (C. Mills, unpubl. data).
P. flavicirrata does not feed exclusively on larvae, however, and the
presence of natural background plankton can significantly reduce
predation compared to that observed in filtered seawater (Figure 4).
Instantaneous mortality of marked bivalve larvae by P. flavicirrata was -
.0025 day·l, which could potentially result in a 7% reduction of a veliger
population over a 28 day planktonic period. While this is not as dramatic as
the potential predation by N. scintillans, it is substantial mortality with the
potential to influence the numbers of larvae available for recruitment.
Other studies showing differential mortality with development
revealed higher vulnerability in earlier stages, as opposed to the
vulnerability in later stages observed in our experiments. Early
vulnerability in early stage echinoids was attributed to a lack of the larval
structures and behavior present in later stages (Rumrill et al., 1985;
Pennington et al., 1986). Possible explanations for the reverse pattern
observed with P. flavicirratli and bivalve veligers may be mechanics- or
behavior-based. As a result of increased swimming speed or increased
body diameter, larger bivalve veligers may contact tentacles more
frequently. Another possible explanation of this predation includes a
143
behavioral or mechanical response of the medusa to the presence of larger
veligers which precludes the capture of d-hinge veligers.
Predation in the corrals was variable. While many experiments
showed little or no predation on benthic invertebrate larvae, the presence of
N. scintillans and P. flavicirrata in corrals resulted in limited to high
predation. Corral and laboratory experiments show that mortality can vary
depending on the interactions between particular species. Larval mortality
may also depend on developmental stage or size. The diversity of conditions
and species that can influence mortality and cause variation in risk
support the idea that the timing of larval release is an important part of
invertebrate reproductive ecology and life history evolution (Giese and
Kanatani, 1987; Morgan, 1995b). There are likely complex interactions
between benthic adults, their planktonic larvae, other components of the
plankton, and physical parameters. Loss of propagules may result from
some combination of fertilization failure, low food availability, an adverse
physical environment, predation, and unfavorable transport. These same
factors may also indirectly affect larvae by affecting predators, competitors,
and food sources. Deciphering the subtleties of these interactions promises
to yield new and exciting discoveries and avenues of research.
Planktonic predationvmay not always explain the substantial loss
which occurs between spawning and benthic recruitment. The fact
remains, however, that benthic invertebrates spawn vast numbers of
larvae, only a few of which recruit to the adult population. Ifwe accede that
there are cases where planktonic predation is not responsible, what is then
144
the source of great loss? Fertilization failure potentially results in huge
losses of spawned eggs. Given the vast number of eggs typically spawned,
however, even low fertilization rates produce many planktonic offspring.
Some larvae may starve in the plankton, though investigations with soft-
bodied larvae suggest that the ability to uptake DOM can prevent starvation
(Olson and Olson, 1989). Lethal temperature extremes or fluctuations may
directly or indirectly contribute to the mortality of larvae (Pechenik, 1987;
Morgan, 1995b). Lack of a suitable settlement site can potentially prevent
entire populations from recruiting (Jackson and Strathmann, 1981). For
example, if the larvae of an intertidal species are transported permanently
away from shore, then none of the population will find a proper place to
settle. Finally, substantial post-settlement mortality may occur on the
benthos before juveniles are large enough to be scored as recruits to the
benthic population.
It is unlikely that anyone phenomenon is always or solely
responsible for larval mortality. Given the complexities of interaction
between the benthos, other planktonic species, and ocean physics,
investigators must conceive holistic approaches for measuring mortality.
We determined survivorship in near-natural plankton assemblages using
in situ corrals and methods'-\vhich allowed us to identify predators. In our
investigation, while N. scintillans was a significant predator in 2 of 10
experiments, the majority of observations showed that veliger and pluteus
larvae suffered little or no predation. Thus, planktonic predation may not
always be a major source of larval mortality.
145
Selected References
Ayers, J. C. 1956. Population dynamics of the marine clam, Mya arenaria.
Limnology and Oceanography 1:26-34.
Babcock, R. C., C. N. Mundy, J. Keesing, and J. Oliver. 1992. Predictable
and unpredictable spawning events: in situ behavioural data from free-
spawning coral reef invertebrates. Invertebrate Reproduction and
Development 22:213-228.
Bailey, K. M., and J. Yen. 1983. Predation by a carnivorous marine
copepod, Euchaeta elongata Esterly, on eggs and larvae of the Pacific
hake, Merluccius productus. Journal of Plankton Research 5:71-82.
Bailey, J. E., B. L. Wing, and C. R. Mattson. 1975. Zooplankton abundance
and feeding habits of fry of pink salmon, Oncorhynchus gorbuscha, and
chum salmon, Oncorhynchus keta, in traitors cove, Alaska, with
speculations on the carrying capacity of the area. Fishery Bulletin 73:846-
861.
Bingham, B. L., and L. J. Walters. 1989. Solitary ascidians as predators of
invertebrate larvae: Evidence from gut analyses and plankton samples.
Journal of Experimental Marine Biology and Ecology 131:147-159.
Boidron-Metairon,1. F. 1995. Larval nutrition. Pages 223-248 in L.
McEdward, editor. Ecology of Marine Invertebrate Larvae. CRC Press,
Boca Raton, Florida, USA.
Brazeau, D. A., and H. R. Lasker. 1992. Reproductive success in the
Caribbean octocoral Briareum asbestinum. Marine Biology 114:157-163.
Burrell, V. G. J., and W. A. Van Engel. 1976. Predation by and distribution
of a ctenophore, Mnemiopsis leidyi A. Agassiz, in the York River
estuary. Estuarine and Coastal Marine Science 4:23.5-242.
Cameron, R. A., and S. S. Rumrill. 1982. Larval abundance and
recruitment of the sand dollar Dendraster excentricus in Monterey Bay,
California, USA. Marine'Biology 71:197-202.
Capriulo, G. M., E. B. Sherr, and B. F. Sherr. 1991. Trophic behaviour and
related community feeding activities of heterotrophic marine protists.
Pages 219-265 in P. C. Reid, editor. Protozoa and their role in marine
processes, G 25. Springer-Verlag, Berlin, Germany.
Carriker, M. R. 1951. Ecological observations on the distribution of oyster
larvae in New Jersey estuaries. Ecological Monographs 21:19-38.
146
Chandy, S. T., and C. H. Greene. 1995. Estimating the predatory impact of
gelatinous zooplankton. Limnology and Oceanography 40:947-955.
Chia, F.-S., and C. W. Walker. 1991. Echinodermata: Asteroidea. Pages
301-353 in A. C. Giese, J. S. Pearse and V. B. Pearse, editors.
Reproduction of Marine Invertebrates V. VI Echinoderms and
Lophophorates. The Boxwood Press, Pacific Grove, California, USA.
Connell, J. H. 1985. The consequences of variation in initial settlement vs.
post-settlement mortality in rocky intertidal communities. The Journal
of Experimental Marine Biology and Ecology 93:11-45.
Cowan, J. H., Jr., E. D. Houde, and J.E. Purcell. 1994. Predator-prey
interactions in the estuarine plankton community of the Chesapeake
Bay, USA. 37th Conference of the International Association for Great
Lakes Research and Estuarine Research Federation: Program and
Abstracts, IAGLR, Buffalo, New York 166:166.
Cronin, T., and B. Forward. 1980. Effects of Starvation on Phototaxis and
Swimming of Larvae of the Crab Rhithropanopeus harrisii. Biological
Bulletin 158:283-294.
DeMott, W. R., and M. D. Watson. 1991. Remote detection of algae by
copepods: Responses to algal size, odors and motility. Journal of
Plankton Research 13:1203-1222.
Emery, A. R. 1973. Comparative ecology and functional osteology of fourteen
species of damselfish (Pisces: Pomacentridae) at Alligator Reef, Florida
Keys. Bulletin of Marine Science 23:649-770.
Emlet, R. B. 1986. Larval production, dispersal, and growth in a fjord: a
case study on larvae of the sand dollar Dendraster excentricus. Marine
Ecology Progress Series 31:245-254.
Enomota, Y. 1956. On the occurrence and the food of Noctiluca scintillans
(Macartney) in the waters adjacent to the west coast of Kyushu, with
special reference to the possibility of the damage caused to fish eggs by
that plankton. Bulletin ofthe Japanese Society of Scientific Fisheries
22:82-89.
Farrell, T. M., D. Bracher, and J. Roughgarden. 1991. Cross-shelf
transport causes recruitment to intertidal populations in central
California. Limnology and Oceanography 36:279-288.
147
Feigenbaum, D., and M. R. Reeve. 1977. Prey detection in the
Chaetognatha: response to a vibrating probe and experimental
determination of attack distance in large aquaria. Limnology and
Oceanography 22:1052-1058.
Feigenbaum, D. L., and R. C. Maris. 1984. Feeding in the Chaetognatha.
Oceanography and Marine Biology Annual Review 22:343-392.
Forward, R. B., Jr., and T. W. Cronin. 1980. Tidal rhythms of activity and
phototaxis of an estuarine crab larva. Biological Bulletin 158:295-303.
Gaines, S., S. Brown, and J. Roughgarden. 1985. Spatial variation in larval
concentrations as a cause of variation in settlement for the barnacle,
Balanus glandula. Oecologia 67:267-272.
Galtsoff, P. S. 1964. The American oyster Crassostrea virginica Gmelin.
Fishery Bulletin 64:1-480.
Gerritsen, J., and J. R. Strickler. 1977. Encounter probabilities and
community structure in zooplankton: a mathematical model. Journal of
the Fisheries Research Board of Canada 34:73-82.
Gibb, J. A. 1962. Tinbergen's hypothesis of the role of specific search
images. Ibis 104:106-111.
Giese, A. C., and H. Kanatani. 1987. Maturation and spawning. Pages 252-
313 in A. C. Giese, J. S. Pearse, and V. B. Pearse, editors. Reproduction
of Marine Invertebrates IX. General Aspects: Seeking Unity in Diversity.
The Boxwood Press, Pacific Grove, California, USA.
Giguere, L. A., and T. G. Northcote. 1987. Ingested prey increase risks of
visual predation in transparent Chaoborus larvae. Oecologia. 73:48-52.
Giske, J., D. L. Aksnes, and O. Fiksen. 1994. Visual predators,
environmental variables and zooplankton mortality risk. Vie Milieu
44:1-9.
Glasgow, H. B. J., J. M. Bwkholder, D. E. Schmechel, P. A. Telster, and P.
A. Rublee. 1995. Insidious effects of a toxic estuarine dinoflagellate on
fish survival and human health. Journal of Toxicology and
Environmental Health 46:501-505.
Glynn, P. W. 1973. Ecology of a Caribbean coral reef. The Porites reef-flat
biotope: Part II. Plankton community with evidence for depletion.
Marine Biology 22:1-21.
148
Greene, C. H. 1985. Planktivore functional groups and patterns of prey
selection in pelagic communities. Journal of Plankton Research 7:35-40.
Hansen, B., P. J. Hansen, and T. G. Nielsen. 1991. Effects of large
nongrazable particles on clearance and swimming behaviour of
zooplankton. Journal of Experimental Marine Biology and Ecology
153:257-269.
Hattori, S. 1962. Predatory activity ofNoctiluca on anchovy eggs. Bulletin of
the Tokai Regional Fisheries Research Laboratory 9:211-220.
Hidu, H., and H. H. Haskin. 1978. Swimming speeds of oyster larvae
Crassostrea virginica in different salinities and temperatures. Estuaries
1:252-255.
Horridge, G. A., and P. S. Boulton. 1967. Prey detection by Chaetognatha via
a vibration sense. Proceedings of the Royal Society of London Series B.
168:413-419.
Jackson, G. A., and R. R. Strathmann. 1981. Larval mortality from offshore
mixing as a link between precompetent and competent periods of
development. The American Naturalist 118:16-26.
Jeong, H. J. 1994. Predation by the heterotrophic dinoflagellate
Protoperidinium cf. divergens on copepod eggs and early naupliar
stages. Marine Ecology Progress Series. 114:1-2.
Johnson, K. B., and A. L. Shanks. 1997. The importance of prey densities
and background plankton in studies of predation on invertebrate larvae.
Marine Ecology Progress Series 158:293-296.
Johnson, K. B., and L. A. Brink. 1998. Predation on bivalve veligers by
polychaete larvea. Biological Bulletin In press.
Kesseler, H. 1966. Beitrag zur Kenntnis der chemischen und
physikalischen Eigenschaften des Zellsaftes von Noctiluca miliaris.
Veroff. Institut Meeresforsch. Bremerh. Sonderband 2:357-368.
Kimor, B. 1979. Predation by Noctiluca miliaris Souriray on Acartia tonsa
Dana eggs in the inshore waters of southern California. Limnology and
Oceanography 24:568-572.
Ki~rboe, T., and J. Titelman. 1998. Feeding, prey selection and prey
encounter mechanisms in the heterotrophic dinoflagellate Noctiluca
scintillans. Journal of Plankton Research In press.
149
Kirchner, M., G. Sahling, G. Uhlig, W. Gunkel, and K. W. Klings. 1996.
Does the red tide-forming dinoflagellate Noctiluca scintillans feed on
bacteria? Sarsia 81:45-55.
Knudsen, J. W. 1960. Reproduction, life history and larval ecology of the
California Xanthidae, the pebble crabs. Pacific Science 14:3-17.
Konstantinova, M. 1. 1966. Characteristics of movement of pelagic larvae of
marine invertebrates. Dokl. Akad. Nauk. SSSR 170:726-729.
Konstantinova, M. 1. 1969. Movement of polychaete larvae. Dokl. Akad.
Nauk. SSSR 188:942-945.
Larson, E. T., and A. L. Shanks. 1996. Consumption of marine snow by two
species of juvenile mullet and its contribution to their growth. Marine
Ecology Progress Series 130:19-28.
Latz, M. I., and R. B. J. Forward. 1977. The effect of salinity upon
phototaxis and geotaxis in a larval crustacean. Biological Bulletin
153:163-179.
Lebour, M. V. 1922. The food of planktonic organisms. Journal of the
Marine Biological Association of the United Kingdom 12:644-677.
Levin, L. A., and T. S. Bridges. 1995. Pattern and diversity in reproduction
and development. Pages 1-48 in L. McEdward, editor. Ecology of Marine
Invertebrate Larvae. CRC Press, Boca Raton, Florida, USA.
McCormick, J. M. 1969. Trophic relationships of hydromedusae in Yaquina
Bay, Oregon. Northwest Science 43:207-214.
Miller, B. 1995. Larval abundance and early juvenile recruitment of
echinoids, asteroids, and holothuroids on the Oregon coast. Masters
Thesis, University of Oregon, Eugene, Oregon, USA.
Mook, J. H., L. J. Mook, and H. S. Heikens. 1960. Further evidence for the
role of "searching images" in the hunting behavior of titmice. Arch.
Need. Zool. 13:448-465.
Morgan, S. G. 1987. Morphological and behavioral antipredatory
adaptations of decapod zoeae. Oecologia 73:393-400.
Morgan, S. G. 1989. Adaptive significance of spination in estuarine crab
zoeae. Ecology 70:464-482.
Morgan, S. G. 1995a. Life and Death in the Plankton: Larval Mortality and
Adaptation. Pages 279-322 in L. McEdward, editor. Ecology of Marine
Invertebrate Larvae. CRC Press, Boca Raton, Florida, USA.
150
Morgan, S. G. 1995b. The Timing of Larval Release. Pages 157-191 in L.
McEdward, editor. Ecology of Marine Invertebrate Larvae. CRC Press,
Boca Raton, Florida, USA.
Morris, R. H., D. P. Abbott, and E. Haderlie. 1980. Intertidal Invertebrates
of California. Stanford University Press, Stanford, California, USA.
Murton, R. K. 1971. The significance of a specific search image in the
feeding behaviour of the wood-pigeon. Behaviour 40:10-42.
Nelson, T. C. 1925. On the occurrence and food habits ofctenophores in New
Jersey inland coastal waters. Biological Bulletin 48:92-111.
Olson, R. R., and M. H. Olson. 1989. Food limitation of planktotrophic
marine invertebrate larvae: does it control recruitment success? Annual
Review of Ecology and Systematics 20:225-247.
Omori, M., and W. M. Hamner. 1982. Patchy distribution of zooplankton:
Behavior, population assessment and sampling problems. Marine
Biology 72:193-200.
Omori, M., and T. Ikeda. 1984. Methods in Marine Zooplankton Ecology,
John Wiley & Sons, New York, New York, USA.
Pechenik, J. A. 1987. Environmental influences on larval survival and
development. Pages 551-595 in A. C. Giese, J. S. Pearse and V. B. Pearse,
editors. Reproduction of Marine Invertbrates IX General Aspects:
Seeking Unity in Diversity. Blackwell Scientific Publications and the
Boxwood Press, Palo Alto and Pacific Grove, California, USA.
Pennington, J. T., and F. S. Chiao 1984. Morphological and behavioral
defenses of trochophore larvae of Sabellaria cementarium (Polychaeta)
against four planktonic predators. Biological Bulletin 167:168-175.
Pennington, J. T., S. S. Rumrill, and F. S. Chiao 1986. Stage-specific
predation upon embryos and larvae of the pacific sand dollar, Dendraster
excentricus, by 11 species of common zooplanktonic predators. Bulletin of
Marine Science 39:234-240.
v
Perron, F. E. 1986. Life history consequences of differences in
developmental mode among gastropods in the genus Conus. Bulletin of
Marine Science 39:485-497.
Phillips, N. E., and B. Pernet. 1996. Capture of large particles by
suspension-feeding scaleworm larvae (Polychaeta: Polynoidae).
Biological Bulletin 191:199-208.
151
Prasad, R. R. 1958. A note on the occurrence and feeding habits of Noctiluca
and their effects on the plankton community and fisheries. Proceedings
of the Indian Academy of Sciences B 67:331-337.
Purcell, J. E. 1982. Feeding and growth of the siphonophore Muggiaea
atlantica (Cunningham 1893). Journal of Experimental Marine Biology
and Ecology 62:39-54.
Purcell, J. E. 1990. Soft-bodied zooplankton predators and competitors of
larval herring (Clupea harengus pallasi) at herring spawning grounds
in British Columbia. Canadian Journal of Fisheries and Aquatic
Sciences 47:505-515.
Purcell, J. E., D. A. Nemazie, S. E. Dorsey, E. D. Houde, and J. C. Gamble.
1994. Predation mortality of bay anchovy Anchoa mitchlli eggs and larvae
due to scyphomedusae and ctenophores in Chesapeake Bay. Marine
Ecology Progress Series 114:47-58.
Reeve, M. R. 1977. The effect of laboratory conditions on the extrapolation of
experimental measurements to the ecology of marine zooplankton. 5. A
review. Proceedings of the Symposium on Warm Water Zooplankton
(NIO) pp. 528-537.
Reeve, M. R. 1980. Comparative experimental studies on the feeding of
chaetognaths and ctenophores. Journal of Plankton Research 2:381-393.
Reeve, M. R. 1981. Large cod-end reservoirs as an aid to the live collection of
delicate zooplankton. Limnology and Oceanography 26:577-580.
Roughgarden, J., S. Gaines, and Y. Iwasa. 1984. Dynamics and evolution
of marine populations with pelagic larval dispersal. Pages 111-128 in R.
M. May, editor. Exploitation of Marine Communities. Springer-Verlag,
New York, New York, USA.
Roughgarden, J., J. T. Pennington, D. Stoner, S. Alexander, and K. Miller.
1991. Collisions of upwelling fronts with the intertidal zone: the cause of
recruitment pulses in barnacle populations of central California. Acta
Oecologica 12:35-51.
Rowley, R. J. 1993. Development of a prototype system for marking and
tracking marine larvae. Abstract from The First Larval Biology
Conference, SUNY Stonybrook, New York, USA.
Rumrill, S. S. 1987. Differential Predation upon embryos and larvae of
Pacific echinoderms. Ph.D. Thesis, University of Alberta, Edmonton,
Canada.
152
Rumrill, S. S. 1990. Natural mortality of marine invertebrate larvae.
Ophelia 32:163-198.
Rumrill, S. S., J. T. Pennington, and F. S. Chiao 1985. Differential
susceptibility of marine invertebrate larvae: Laboratory predation of sand
dollar, Dendraster excentricus (Eschscholtz), embryos and larvae by
zoeae of the red crab, Cancer productus Randall. Journal of
Experimental Marine Biology and Ecology 90:193-208.
Sebens, K. P., and M. A. R. Koehl. 1984. Predation on zooplankton by the
benthic anthozoans Alcyonium siderium (Alcyonacea) and Metridium
senile (Actiniaria) in the New Enland subtidal. Marine Biology 81:255-
271.
Strathmann, M. 1987. Reproduction and Development of Marine
Invertebrates of the Northern Pacific Coast. University of Washington
Press, Seattle, Washington, USA.
Strathmann, R. R. 1974. The spread of sibling larvae of sedentary marine
invertebrates. The American Naturalist 108:29-44.
Strathmann, R. R. 1985. Feeding and nonfeeding larval development and
life-history evolution in marine invertebrates. Annual Review of Ecology
and Systematics 16:339-361.
Strathmann, R. R. 1990. Why life histories evolve differently in the sea.
American Zoologist 30:197-207.
Sulkin, S. D. 1973. Depth regulation of crab larvae in the absence of light.
Journal of Experimental Marine Biology and Ecology 13:73-82.
Sulkin, S. D. 1975. The influence of light in the depth regulation of crab
larvae. Biological Bulletin 148:333-343.
Thorson, G. 1946. Reproduction and larval development of danish marine
bottom invertebrates. Meddr Komm Danm Fisk -og Havunders serie
Plankton 4:1-523.
Thorson, G. 1950. Reproductive and larval ecology of marine bottom
invertebrates. Biological R'eview 25:1-45.
Thorson, G. 1966. Some factors influencing the recruitment and
establishment of marine benthic communities. Netherlands Journal of
Sea Research 3:267-293.
Tinbergen, L. 1960. The natural control of insects in pinewoods, 1: factors
influencing the intensity of predation by songbirds. Arch. Neerl. Zool.
13:266-336.
153
Toonen, R. J., and F. S. Chiao 1993. Limitations of laboratory assessments
of coelenterate predation: container effects on the prey selection of the
Limnomedusa, Proboscidactyla flavicirrata (Brandt). Journal of
Experimental Marine Biology and Ecology 167:215-235.
Yen, J. 1987. Predation by a carnivorous marine copepod, Euchaeta
norvegica Boeck, on eggs and larvae of the North Atlantic cod Gadus
morhua L. Journal of Experimental Marine Biology and Ecology. 112:283-
296.
Yen, J., and N. T. Nicoll. 1990. Setal array on the first antennae of a
carnivorous marine copepod, Euchaeta norvegica. Journal of
Crustacean Biology 10:218-224.
Young, C. M. 1989. Larval depletion by ascidians has little effect on
settlement of epifauna. Marine Biology 102:481-489.
Young, C. M., and N. J. Gotelli. 1988. Larval predation by barnacles: effects
on patch colonization in a shallow subtidal community. Ecology 69:624-
634.
Zimmerman, S. T. 1972. Seasonal Succession of Zooplankton Populations in
Two Dissimilar Marine Embayments on the Oregon Coast. Ph.D. Thesis,
Oregon State University, Oregon, USA.
154
CHAPTER VI
CONCLUDING SUMMARY
This doctoral dissertation research investigated predation on
planktonic invertebrate larvae under the most natural conditions possible
in the laboratory and in the field. Laboratory experiments manipulated
prey density and the presence of background plankton to determine their
effect on predation rates. Results indicate that most predators examined do
not prey on selected larvae when conditions are closest to natural (i.e.,
natural prey densities with background plankton present). Comparisons
with unnaturally high prey density and filtered seawater treatments
suggest that prey density and background plankton can strongly influence
the outcome of predator-prey experiments. Therefore, laboratory
experiments are more applicable to nature when natural prey densities and
background plankton are used.
In situ experiments with near-natural plankton assemblages
usually yielded low predation on marked echinoid, bivalve, and gastropod
larvae. One predator of bivalve veligers, shown in two experiments to
yo
consume significant numbers of larvae, was the heterotrophic
dinoflagellate Noctiluca scintillans. Other potential predators, none of
which heavily impacted marked larval populations, included a variety of
large crustaceans (e.g., copepods, amphipods, and zoeae), various
155
polychaetes, chaetognaths, fish, ctenophores, and hydromedusae.
Recovery of marked larvae, whether unconsumed or in predator guts, was
usually 100%. Predator gut contents indicate that crustaceans, including
crustacean larvae, are the most common prey of many predators.
Juveniles of the threespine stickleback Gasterosteus aculeatus consumed
many crustaceans and seem to selectively feed on the nauplii of barnacles.
While their impact on the wild nauplius populations captured in corrals
was heavy, short residence times of G. aculeatus should greatly reduce the
observed impact in nature.
Field observations and experiments provide a direct assessment of
predation's importance for larvae in an experimental assemblage. The
following strengths of experimental design contribute to the field data's
value for examining predation as an important source of larval invertebrate
mortality. Larval species examined as prey include 3 species of marked
larvae and several other species of wild larvae representing 4 phyla. The
data were collected from interactions with captured natural assemblages,
each with a diversity of potential predators. Initial densities of marked
larvae were known, allowing more powerful analysis of observations.
Predators and prey interacted at natural densities, eliminating the
possibility of artifactual bel1avior induced by abnormal densities. Marked
larvae were easily visible in the guts of predators. Corral volumes far
exceeded the practical capacity of laboratory containers and reduced
artifacts resulting from small volumes. Finally, corral samples were
collected and fixed immediately at the close of experiments, minimizing the
156
possibility of predation artifacts in the concentrated sample. These in situ
studies are valuable for examining predation and mortality on planktonic
invertebrate larvae.
Low predation rates observed in near-natural plankton assemblages
may result from low predator-prey encounters. However, model estimates
of encounter rates within the corrals indicate that numerous predator-prey
encounters should have occurred. One explanation for low predation in
spite of encounter estimates is the potential influence of background
plankton on predation. Background plankton may serve as substitute food
or reduce encounters. Corral manipulations of prey density and
background plankton support laboratory findings that natural prey
densities and background plankton can reduce or eliminate predation.
While N. scintillans was a significant predator in 2 of 10 field experiments,
the majority of observations from both the field and the laboratory showed
that veliger and pluteus larvae suffered little or no predation under near-
natural conditions. Thus, planktonic predation may not always be a major
source of larval mortality.
157
BIBLIOGRAPHY
Alldredge, A. L., and W. M. Hamner. 1980. Recurring aggregation of
zooplankton by a tidal current. Estuarine and Coastal Marine Science
10:31-37.
Ayers, J. C. 1956. Population dynamics of the marine clam, Mya arenaria.
Limnology and Oceanography 1:26-34.
Babcock, R. C., C. N. Mundy, J. Keesing, and J. Oliver. 1992. Predictable
and unpredictable spawning events: in situ behavioural data from free-
spawning coral reef invertebrates. Invertebrate Reproduction and
Development 22:213-228.
Bailey, J. E., B. L. Wing, and C. R. Mattson. 1975. Zooplankton abundance
and feeding habits of fry of pink salmon, Oncorhynchus gorbuscha, and
chum salmon, Oncorhynchus keta, in traitors cove, Alaska, with
speculations on the carrying capacity of the area. Fishery Bulletin 73:846-
861.
Bailey, K. M., and R. S. Battey. 1983. A laboratory study ofpredation by
Aurelia aurita on larval herring, Clupea harengus: experimental
observations compared with model predictions. Marine Biology 72:295-
302.
Bailey, K. M., and J. Yen. 1983. Predation by a carnivorous marine
copepod, Euchaeta elongata Esterly, on eggs and larvae of the Pacific
hake, Merluccius productus. Journal of Plankton Research 5:71-82.
Bayne, B. L. 1965. Growth and the delay of metamorphosis of the larvae of
Mytilus edulis (L.). Ophelia 2:1-47.
Beaumont, A. R., and M. D. Budd. 1982. Delayed growth of mussel (Mytilus
edulis ) and scallop (Pecten maximus ) veligers at low temperatures.
Marine Biology 71:97-1mr.
Bhaud, M., and C. Cazaux. 1987. Description and identification of
polychaete larvae; their implications in current biological problems.
Oceanis 13:596-753.
Bingham, B. L., and L. J. Walters. 1989. Solitary ascidians as predators of
invertebrate larvae: Evidence from gut analyses and plankton samples.
Journal of Experimental Marine Biology and Ecology 131:147-159.
158
Boidron-Metairon, I. F. 1995. Larval nutrition. Pages 223-248 in L.
McEdward, editor. Ecology of Marine Invertebrate Larvae. CRC Press,
Boca Raton, Florida, USA.
Bray, B. M. 1953. Sea-water discoloration by living organisms. New
Zealand Journal of Science and Technology B34:393-407.
Brazeau, D. A., and H. R. Lasker. 1992. Reproductive success in the
Caribbean octocoral Briareum asbestinum. Marine Biology 114:157-163.
Brink, L.A. 1997. Cross-shelf transport of planktonic larvae of inner shelf
benthic invertebrates. M.S. Thesis, University of Oregon, Eugene,
Oregon, USA.
Burrell, V. G. J., and W. A. Van Engel. 1976. Predation by and distribution
of a ctenophore, Mnemiopsis leidyi A. Agassiz, in the York River
estuary. Estuarine and Coastal Marine Science 4:235-242.
Cameron, R. A., and S. S. Rumrill. 1982. Larval abundance and
recruitment of the sand dollar Dendraster excentricus in Monterey Bay,
California, USA. Marine Biology 71:197-202.
Capriulo, G. M., E. B. Sherr, and B. F. Sherr. 1991. Trophic behaviour and
related community feeding activities of heterotrophic marine protists.
Pages 219-265 in P. C. Reid, editor. Protozoa and their role in marine
processes, G 25. Springer-Verlag, Berlin, Germany.
Carriker, M. R. 1951. Ecological observations on the distribution of oyster
larvae in New Jersey estuaries. Ecological Monographs 21:19-38.
Chandy, S. T., and C. H. Greene. 1995. Estimating the predatory impact of
gelatinous zooplankton. Limnology and Oceanography 40:947-955.
Chia, F.-S., and C. W. Walker. 1991. Echinodermata: Asteroidea. Pages
301-353 in A. C. Giese, J. S. Pearse and V. B. Pearse, editors.
Reproduction of Marine Invertebrates V. VI Echinoderms and
Lophophorates. The Boxwood Press, Pacific Grove, California, USA.
Connell, J. H. 1985. The cotisequences of variation in initial settlement vs.
post-settlement mortality in rocky intertidal communities. The Journal
of Experimental Marine Biology and Ecology 93:11-45.
Cowan, J. H., Jr., E. D. Houde, and J.E. Purcell. 1994. Predator-prey
interactions in the estuarine plankton community of the Chesapeake
Bay, USA. 37th Conference of the International Association for Great
Lakes Research and Estuarine Research Federation: Program and
Abstracts, IAGLR, Buffalo, New York 166:166.
159
Cronin, T., and B. Forward. 1980. Effects of Starvation on Phototaxis and
Swimming of Larvae of the Crab Rhithropanopeus harrisii. Biological
Bulletin 158:283-294.
Daro, M. H., and P. Polk. 1973. The autecology of Polydora ciliata along the
Belgian coast. Netherlands Journal of Sea Research 6:130-140.
De Burgh, M. E., and R. D. Burke. 1983. Uptake of dissolved amino acids by
embryos and larvae of Dendraster excentricus (Eschscholtz)
(Echinodermata: Echinoidea). Canadian Journal of Zoology 61:349-354.
DeMott, W. R., and M. D. Watson. 1991. Remote detection of algae by
copepods: Responses to algal size, odors and motility. Journal of
Plankton Research 13:1203-1222.
Emery, A. R. 1973. Comparative ecology and functional osteology of fourteen
species of damselfish (Pisces: Pomacentridae) at Alligator Reef, Florida
Keys. Bulletin of Marine Science 23:649-770.
Emlet, R. B. 1986. Larval production, dispersal, and growth in a fjord: a
case study on larvae of the sand dollar Dendraster excentricus. Marine
Ecology Progress Series 31:245-254.
Enomota, Y. 1956. On the occurrence and the food of Noctiluca scintillans
(Macartney) in the waters adjacent to the west coast of Kyushu, with
special reference to the possibility of the damage caused to fish eggs by
that plankton. Bulletin of the Japanese Society of Scientific Fisheries
22:82-89.
Evans, G. T. 1989. The encounter speed of moving predator and prey.
Journal of Plankton Research 11:415-417.
Farrell, T. M., D. Bracher, and J. Roughgarden. 1991. Cross-shelf
transport causes recruitment to intertidal populations in central
California. Limnology and Oceanography 36:279-288.
Feigenbaum, D., and M. R. Reeve. 1977. Prey detection in the
Chaetognatha: response to a vibrating probe and experimental
determination of attack"'distance in large aquaria. Limnology and
Oceanography 22:1052-1058.
Feigenbaum, D. L., and R. C. Maris. 1984. Feeding in the Chaetognatha.
Oceanography and Marine Biology Annual Review 22:343-392.
Forward, R. B., Jr., and T. W. Cronin. 1980. Tidal rhythms of activity and
phototaxis of an estuarine crab larva. Biological Bulletin 158:295-303.
160
Gaines, S., S. Brown, and J. Roughgarden. 1985. Spatial variation in larval
concentrations as a cause of variation in settlement for the barnacle,
Balanus glandula. Oecologia 67:267-272.
Gallager, S. M. 1988. Visual observations of particle manipulation during
feeding in larvae of a bivalve mollusc. Bulletin of Marine Science 43:344-
365.
Galtsoff, P. S. 1964. The American oyster Crassostrea virginica Gmelin.
Fishery Bulletin 64:1-480.
George, D. G., and R. W. Edwards. 1973. Daphnia distribution within
langmuir circulations. Limnology and Oceanography 18:798-800.
Gerritsen, J., and J. R. Strickler. 1977. Encounter probabilities and
community structure in zooplankton: a mathematical model. Journal of
the Fisheries Research Board of Canada 34:73-82.
Gibb, J. A. 1962. Tinbergen's hypothesis of the role of specific search
images. Ibis 104:106-111.
Giese, A. C., and H. Kanatani. 1987. Maturation and spawning. Pages 252-
313 in A. C. Giese, J. S. Pearse, and V. B. Pearse, editors. Reproduction
of Marine Invertebrates IX. General Aspects: Seeking Unity in Diversity.
The Boxwood Press, Pacific Grove, California, USA.
Giguere, L. A., A. Delage, L. M. Dill, and J. Gerritsen. 1982. Predicting
Encounter Rates for Zooplankton: A Model Assuming a Cylindrical
Encounter Field. Canadian Journal of Fisheries and Aquatic Sciences
39:237-242.
Giguere, L. A., and T. G. Northcote. 1987. Ingested prey increase risks of
visual predation in transparent Chaoborus larvae. Oecologia. 73:48-52.
Giske, J., D. L. Aksnes, and O. Fiksen. 1994. Visual predators,
environmental variables and zooplankton mortality risk. Vie Milieu
44:1-9.
Glasgow, H. B. J., J. M. Burkholder, D. E. Schmechel, P. A. Telster, and P.
A. Rublee. 1995. Insidious effects of a toxic estuarine dinoflagellate on
fish survival and human health. Journal of Toxicology and
Environmental Health 46:501-505.
Glynn, P. W. 1973. Ecology of a Caribbean coral reef. The Porites reef-flat
biotope: Part II. Plankton community with evidence for depletion.
Marine Biology 22:1-21.
161
Green, E. P., and M. J. Dagg. 1997. Mesozooplankton associations with
medium to large marine snow aggregates in the northern Gulf of
Mexico. Journal of Plankton Research 19:435-447.
Greene, C. H. 1985. Planktivore functional groups and patterns of prey
selection in pelagic communities. Journal of Plankton Research 7:35-40.
Hamner, W. H., and J. H. Carleton. 1979. Copepod swarms: attributes and
role in coral reef ecosystems. Limnology and Oceanography 24:1-14.
Hansen, B., P. J. Hansen, and T. G. Nielsen. 1991. Effects of large
nongrazable particles on clearance and swimming behaviour of
zooplankton. Journal of Experimental Marine Biology and Ecology
153:257-269.
Harms, J. 1984. Influence of water temperature on larval development of
Elminius modestus and Semibalanus balanoides (Crustacea,
Cirripedia). Helgolander wiss. Meeresunters 38:123-134.
Hattori, S. 1962. Predatory activity ofNoctiluca on anchovy eggs. Bulletin of
the Tokai Regional Fisheries Research Laboratory 9:211-220.
Hidu, H., and H. H. Haskin. 1978. Swimming speeds of oyster larvae
Crassostrea virginica in different salinities and temperatures. Estuaries
1:252-255.
Holling, C. S. 1959. The components of predation as revealed by a study of
small mammal predation of the European pine sawfly. Canadian
Entomology 91:293-320.
Horridge, G. A., and P. S. Boulton. 1967. Prey detection by Chaetognatha via
a vibration sense. Proceedings of the Royal Society of London Series B.
168:413-419.
Jackson, G. A., and R. R. Strathmann. 1981. Larval mortality from offshore
mixing as a link between precompetent and competent periods of
development. The American Naturalist 118:16-26.
Jaeckle, W. B., and D. T. Manahan. 1988. Feeding in lecithotrophic larvae
of Haliotus rufescens: uptake of dissolved amino acids from seawater.
American Zoologist 28:167A.
Jaeckle, W. B., and D. T. Manahan. 1989. Growth and energy balance
during the development of a lecithotrophic molluscan larva (Haliotus
rufescens). Biological Bulletin 177:237-246.
162
Jeong, H. J. 1994. Predation by the heterotrophic dinoflagellate
Protoperidinium cf. divergens on copepod eggs and early naupliar
stages. Marine Ecology Progress Series. 114:1-2.
Johnson, K. B., and A. L. Shanks. 1997. The importance of prey densities
and background plankton in studies of predation on invertebrate larvae.
Marine Ecology Progress Series 158:293-296.
Johnson, K. B., and L. A. Brink. 1998. Predation on bivalve veligers by
polychaete larvea. Biological Bulletin In press.
Kesseler, H. 1966. Beitrag zur Kenntnis der chemischen und
physikalischen Eigenschaften des Zellsaftes von Noctiluca miliaris.
Veroff. Institut Meeresforsch. Bremerh. Sonderband 2:357-368.
Kimor, B. 1979. Predation by Noctiluca miliaris Souriray on Acartia tonsa
Dana eggs in the inshore waters of southern California. Limnology and
Oceanography 24:568-572.
Kingston, P. 1974. Some observations on the effects of temperature and
salinity upon the growth of Cardium edule and Cardium glaucum larvae
in the laboratory. Journal of the Marine Biological Association of the
United Kingdom 54:309-317.
Ki~rboe, T., and J. Titelman. 1998. Feeding, prey selection and prey
encounter mechanisms in the heterotrophic dinoflagellate Noctiluca
scintillans. Journal of Plankton Research In press.
Kirchner, M., G. Sahling, G. Uhlig, W. Gunkel, and K. W. Klings. 1996.
Does the red tide-forming dinoflagellate Noctiluca scintillans feed on
bacteria? Sarsia 81:45-55.
Knudsen, J. W. 1960. Reproduction, life history and larval ecology of the
California Xanthidae, the pebble crabs. Pacific Science 14:3-17.
Konstantinova, M. I. 1966. Characteristics of movement of pelagic larvae of
marine invertebrates. Dokl. Akad. Nauk. SSSR 170:726-729.
Konstantinova, M. I. 1969~'Movementof polychaete larvae. Dokl. Akad.
Nauk. SSSR 188:942-945.
Krebs, J. R., J. T. Erichsen, M. I. Webber, and E. L. Charnov. 1977.
Optimal prey-selection by the great tit (Parus major). Animal Behavior
25:30-38.
163
Kuhl, H. 1974. Uber vorkemmen und nahrung der larven von Magelona
papillicornis O.F. Muller (Polychaeta Sedentaria) im Miindungsgebiet
von elbe, weser und ems. Ber dt wiss Kommn Meeresforschung 23:296-
301.
Larson, E. T., and A. L. Shanks. 1996. Consumption of marine snow by two
species of juvenile mullet and its contribution to their growth. Marine
Ecology Progress Series 130:19-28.
Latz, M. I., and R. B. J. Forward. 1977. The effect of salinity upon
phototaxis and geotaxis in a larval crustacean. Biological Bulletin
153:163-179.
Lebour, M. V. 1922. The food ofplank.tonic organisms. Journal of the
Marine Biological Association of the United Kingdom 12:644-677.
Leighton, D. L. 1974. The influence of temperature on larval and juvenile
growth in three species of southern California abalones. Fishery Bulletin
72:1137-1145.
Levin, L. A., and T. S. Bridges. 1995. Pattern and diversity in reproduction
and development. Pages 1-48 in L. McEdward, editor. Ecology of Marine
Invertebrate Larvae. CRC Press, Boca Raton, Florida, USA.
Levitan, D. R. 1995. The ecology of fertilization in free-spawning
invertebrates. Pages 123-256 in L. McEdward, editor. Ecology of Marine
Invertebrate Larvae. CRC Press, Boca Raton, Florida, USA.
Lima, G. M., and J. A. Pechenik. 1985. The influence of temperature on
growth rate and length of larval life of the gastropod, Crepidula plana
Say. Journal of Experimental Marine Biology and Ecology 90:55-71.
Lucas, A., L. Chebob-Chalabi, and D. A. Aranda. 1986. Transition from
endotrophy to exotrophy in the larvae of Mytilus edulis L. Oceano!. Acta.
9:97-103.
Luo, J., S. A. Brandt, and M. J. Klebasko. 1996. Virtual reality of
plank.tivores: a fish's perspective of prey size selection. Marine Ecology
Progress Series 140:271~83.
Manahan, D. T. 1983. The uptake and metabolism of dissolved amino-acids
by bivalve larvae. Biological Bulletin 164:236-250.
Manahan, D. T., and S. H. Wright. 1991. Uptake of dissolved organic matter
from seawater by marine invertebrates: the "Grover C. Stephens' Era".
American Zoologist 31:3A.
164
McCormick, J. M. 1969. Trophic relationships of hydromedusae in Yaquina
Bay, Oregon. Northwest Science 43:207-214.
Mileikovski, S. A. 1959. Interrelations between the pelagic larvae of
Nephtys ciliata (O.F. Muller), Macoma baltica and Mya arenaria of the
White Sea. Zoologische ZhurnaI38:1889-1891.
Miller, B. 1995. Larval abundance and early juvenile recruitment of
echinoids, asteroids, and holothuroids on the Oregon coast. Masters
Thesis, University of Oregon, Eugene, Oregon, USA.
Mook, J. H., L. J. Mook, and H. S. Heikens. 1960. Further evidence for the
role of "searching images" in the hunting behavior of titmice. Arch.
Neerl. Zool. 13:448-465.
Morgan, S. G. 1987. Morphological and behavioral antipredatory
adaptations of decapod zoeae. Oecologia 73:393-400.
Morgan, S. G. 1989. Adaptive significance of spination in estuarine crab
zoeae. Ecology 70:464-482.
Morgan, S. G. 1995a. Life and Death in the Plankton: Larval Mortality and
Adaptation. Pages 279-322 in L. McEdward, editor. Ecology of Marine
Invertebrate Larvae. CRC Press, Boca Raton, Florida, USA.
Morgan, S. G. 1995b. The Timing of Larval Release. Pages 157-191 in L.
McEdward, editor. Ecology of Marine Invertebrate Larvae. CRC Press,
Boca Raton, Florida, USA.
Morris, R. H., D. P. Abbott, and E. Haderlie. 1980. Intertidal Invertebrates
of California. Stanford University Press, Stanford, California, USA.
Murton, R. K. 1971. The significance of a specific search image in the
feeding behaviour of the wood-pigeon. Behaviour 40:10-42.
Nelson, T. C. 1925. On the occurrence and food habits of ctenophores in New
Jersey inland coastal waters. Biological Bulletin 48:92-111.
O'Brien, W. J., H. I. Browrtlan, and B. I. Evans. 1990. Search strategies of
foraging animals. American Scientist 78:152-160.
Olson, R. R. 1985. In situ culturing of larvae of the crown-of-thorns starfish
Acanthaster planci. Marine Ecology Progress Series 25:207-210.
Olson, R. R. 1987. In situ culturing as a test for the larval starvation
hypothesis for the crown-of-thorns starfish, Acanthaster planci.
Limnology and Oceanography 32:985-904.
165
Olson, R. R., and M. H. Olson. 1989. Food limitation of planktotrophic
marine invertebrate larvae: does it control recruitment success? Annual
Review of Ecology and Systematics 20:225-247.
Omori, M., and W. M. Hamner. 1982. Patchy distribution of zooplankton:
Behavior, population assessment and sampling problems. Marine
Biology 72:193-200.
Omori, M., and T. Ikeda. 1984. Methods in Marine Zooplankton Ecology,
John Wiley & Sons, New York, New York, USA.
PavilIon, J. F. 1976. Etude de l'absorption de quelques substances
organiques dissoutes marquees au 14C et au 3H par les oeufs et les larvae
de certains invertebres marins. Actualites de Biochimie Marine,
Colloque GAB1M - CNRS:211-231.
Pechenik, J. A. 1987. Environmental influences on larval survival and
development. Pages 551-595 in A. C. Giese, J. S. Pearse and V. B. Pearse,
editors. Reproduction of Marine Invertbrates IX General Aspects:
Seeking Unity in Diversity. Blackwell Scientific Publications and the
Boxwood Press, Palo Alto and Pacific Grove, California, USA.
Pennington, J. T., and F. S. Chiao 1984. Morphological and behavioral
defenses of trochophore larvae of Sabellaria cementarium (Polychaeta)
against four planktonic predators. Biological Bulletin 167:168-175.
Pennington, J. T., S. S. Rumrill, and F. S. Chiao 1986. Stage-specific
predation upon embryos and larvae of the pacific sand dollar, Dendraster
excentricus, by 11 species of common zooplanktonic predators. Bulletin of
Marine Science 39:234-240.
Perron, F. E. 1986. Life history consequences of differences in
developmental mode among gastropods in the genus Conus. Bulletin of
Marine Science 39:485-497.
Phillips, N. E., and B. Pernet. 1996. Capture of large particles by
suspension-feeding scaleworm larvae (Polychaeta: Polynoidae).
Biological Bulletin 191:199-208.
>,-
Prasad, R. R. 1958. A note on the occurrence and feeding habits of Noctiluca
and their effects on the plankton community and fisheries. Proceedings
of the Indian Academy of Sciences B 67:331-337.
Purcell, J. E. 1982. Feeding and growth of the siphonophore Muggiaea
atlantica (Cunningham 1893). Journal of Experimental Marine Biology
and Ecology 62:39-54.
166
Purcell, J. E. 1990. Soft-bodied zooplankton predators and competitors of
larval herring (Clupea harengus pallasi) at herring spawning grounds
in British Columbia. Canadian Journal of Fisheries and Aquatic
Sciences 47:505-515.
Purcell, J. E., D. A. Nemazie, S. E. Dorsey, E. D. Houde, and J. C. Gamble.
1994. Predation mortality of bay anchovy Anchoa mitchlli eggs and larvae
due to scyphomedusae and ctenophores in Chesapeake Bay. Marine
Ecology Progress Series 114:47-58.
Reeve, M. R. 1977. The effect of laboratory conditions on the extrapolation of
experimental measurements to the ecology of marine zooplankton. 5. A
review. Proceedings of the Symposium on Warm Water Zooplankton
(NIO) pp. 528-537.
Reeve, M. R. 1980. Comparative experimental studies on the feeding of
chaetognaths and ctenophores. Journal of Plankton Research 2:381-393.
Reeve, M. R. 1981. Large cod-end reservoirs as an aid to the live collection of
delicate zooplankton. Limnology and Oceanography 26:577-580.
Rosenthal, H., and G. Hempel. 1970. Experimental studies in feeding and
food requirements of herring larvae (Clupea harengus L.). pp. 344-364 In
J. H. Steele, editor. Marine Food Chains. Oliver and Boyd, Edinburgh,
Scotland.
Roughgarden, J., S. Gaines, and Y. Iwasa. 1984. Dynamics and evolution
of marine populations with pelagic larval dispersal. Pages 111-128 in R.
M. May, editor. Exploitation of Marine Communities. Springer-Verlag,
New York, New York, USA.
Roughgarden, J., J. T. Pennington, D. Stoner, S. Alexander, and K. Miller.
1991. Collisions of upwelling fronts with the intertidal zone: the cause of
recruitment pulses in barnacle populations of central California. Acta
Oecologica 12:35-51.
Rowley, R. J. 1993. Development of a prototype system for marking and
tracking marine larvae. Abstract from The First Larval Biology
Conference, SUNY Stoil'ybrook, New York, USA.
Rumrill, S. S. 1987. Differential Predation upon embryos and larvae of
Pacific echinoderms. Ph.D. Thesis, University of Alberta, Edmonton,
Canada.
Rumrill, S. S. 1990. Natural mortality of marine invertebrate larvae.
Ophelia 32:163-198.
167
Rumrill, S. S., J. T. Pennington, and F. S. Chiao 1985. Differential
susceptibility of marine invertebrate larvae: Laboratory predation of sand
dollar, Dendraster excentricus (Eschscholtz), embryos and larvae by
zoeae of the red crab, Cancer productus Randall. Journal of
Experimental Marine Biology and Ecology 90:193-208.
Scheltema, R. S. 1967. The relationship of temperature to the larval
development of Nassarius obsoletus (Gastropoda). Biological Bulletin
132:253-265.
Sebens, K. P., and M. A. R. Koehl. 1984. Predation on zooplankton by the
benthic anthozoans Alcyonium siderium (Alcyonacea) and Metridium
senile (Actiniaria) in the New Enland subtidal. Marine Biology 81:255-
271.
Sewell, M. A., and D. R. Levitan. 1992. Fertilization success during a
natural spawning of the dendrochirote sea cucumber Cucumaria
miniata. Bulletin of Marine Science 51:161-166.
Shanks, A. L., and W. G. Wright. 1987. Internal-wave-mediated shoreward
transport of cyprids, megalopae, and gammarids and correlated
longhsore differences in the settling rate of intertidal barnacles. Journal
of Experimental Marine Biology and Ecology 114:1-13.
Shanks, A. L., and K. A. del Carmen. 1997. Larval polychaetes are strongly
associated with marine snow. Marine Ecology Progress Series 154:211-
221.
Shanks, A. L., and K. Walters. 1997. Holoplankton, meroplankton, and
meiofauna associated with marine snow. Marine Ecology Progress
Series 156:75-86.
Smidt, E. L. B. 1951. Animal production in the Danish Waddensea. Meddr
Kommn Danm Fisk- og Havunders 11:151 pp.
Sokal, R. R., and F. J. Rohlf. 1995. Biometry, third edition. WH Freeman
and Company, New York, New York, USA.
Stommel, H. 1949. TrajectOries of small bodies sinking slowly through
convection cells. Journal of Marine Research 8:24-29.
Strathmann, M. 1987. Reproduction and Development of Marine
Invertebrates of the Northern Pacific Coast. University of Washington
Press, Seattle, Washington, USA.
Strathmann, R. R. 1974. The spread of sibling larvae of sedentary marine
invertebrates. The American Naturalist 108:29-44.
168
Strathmann, R. R. 1985. Feeding and nonfeeding larval development and
life-history evolution in marine invertebrates. Annual Review of Ecology
and Systematics 16:339-361.
Strathmann, R. R. 1987. Larval feeding. Pages 465-550 in A. C. Giese, J. S.
Pearse and V. B. Pearse, editors. Reproduction of Marine Invertbrates
IX General Aspects: Seeking Unity in Diversity. Blackwell Scientific
Publications and the Boxwood Press, Palo Alto and Pacific Grove,
California, USA.
Strathmann, R. R. 1990. Why life histories evolve differently in the sea.
American Zoologist 30:197-207.
Sulkin, S. D. 1973. Depth regulation of crab larvae in the absence of light.
Journal of Experimental Marine Biology and Ecology 13:73-82.
Sulkin, S. D. 1975. The influence of light in the depth regulation of crab
larvae. Biological Bulletin 148:333-343.
Thorson, G. 1946. Reproduction and larval development of danish marine
bottom invertebrates. Meddr Komm Danm Fisk -og Havunders serie
Plankton 4:1-523.
Thorson, G. 1950. Reproductive and larval ecology of marine bottom
invertebrates. Biological Review 25:1-45.
Thorson, G. 1966. Some factors influencing the recruitment and
establishment of marine benthic communities. Netherlands Journal of
Sea Research 3:267-293.
Tinbergen, L. 1960. The natural control of insects in pinewoods, 1: factors
influencing the intensity of predation by songbirds. Arch. Neerl. Zool.
13:266-336.
Todd, C. D., M. S. Laverack, and G.A. Boxshall. 1996. Coastal Marine
Zooplankton, second edition. Cambridge University Press, New York,
New York, USA.
Toonen, R. J., and F. S. Chia. 1993. Limitations of laboratory assessments
of coelenterate predation: container effects on the prey selection of the
Limnomedusa, Proboscidactyla flavicirrata (Brandt). Journal of
Experimental Marine Biology and Ecology 167:215-235.
Ueda, H., A. Kuwahara, M. Tanaka, and M. Azeta. 1983. Underwater
observations on copepod swarms in temperate and subtropical waters.
Marine Ecology Progress Series 11:165-171.
169
Wilson, D. P. 1982. The larval development of three species of Magelona
(Polychaeta) from the localities near Plymouth. Journal of the Marine
Biological Association of the United Kingdom 62:385-401.
Yen, J. 1987. Predation by a carnivorous marine copepod, Euchaeta
norvegica Boeck, on eggs and larvae of the North Atlantic cod Gadus
morhua L. Journal of Experimental Marine Biology and Ecology. 112:283-
296.
Yen, J., and N. T. Nicoll. 1990. Setal array on the first antennae of a
carnivorous marine copepod, Euchaeta norvegica. Journal of
Crustacean Biology 10:218-224.
Yokouchi, K. 1991. Seasonal distribution and food habits of planktonic
larvae of benthic polychaetes in Volcano Bay, Southern Hokkaido, Japan.
Ophelia 5:401-410.
. Young, C. M. 1989. Larval depletion by ascidians has little effect on
settlement of epifauna. Marine Biology 102:481-489.
Young, C. M., and N. J. Gotelli. 1988. Larval predation by barnacles: effects
on patch colonization in a shallow subtidal community. Ecology 69:624-
634.
Zimmerman, S. T. 1972. Seasonal Succession of Zooplankton Populations in
Two Dissimilar Marine Embayments on the Oregon Coast. Ph.D. Thesis,
Oregon State University, Oregon, USA.