|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
LIFE-HISTORY CONSEQUENCES OF ARTIFICIAL SELECTION FOR
INCREASED EGG SIZE IN Hydroides elegans (POLYCHAETA: SERPULIDAE)
CECELIA MARIE MILES
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
Cecelia Marie Miles
This work is dedicated to Dr. Larry McEdward. Larry's ability to distill complex topics
down to their essential elements served to focus my efforts as a new graduate student. As
I have advanced in my career, I have come to respect this skill immensely and it is one
that I aspire to emulate. This project was conceived and planned under his guidance and
his influence is present in every aspect.
For insight, guidance, encouragement, and especially for leading by example, I
thank Dr. Marta L. Wayne. I also thank committee members Dr. Colette St. Mary, Dr.
Shirley Baker, and Dr. Gustav Paulay. I am especially grateful to Dr. Ben Bolker for his
clear thinking and concise statistical advice.
I thank Dr. Michael G. Hadfield for his generosity and advice, and everyone at
Kewalo Marine Laboratory in Honolulu, Hawaii for their kokua. I could never have
completed this project without my fellow "worm wrangler" at Kewalo, Sharon Kelly.
I thank the Zoology Department at the University of Florida for financial,
emotional, and spiritual support. I would also like to acknowledge the Auzenne Graduate
Scholars Fellowship from the University of Florida, and the Grinter Fellowship from the
I wish to thank my family and especially my husband, Myron G. Miles, for his
TABLE OF CONTENTS
A C K N O W L E D G M E N T S ................................................................................................. iv
LIST OF TA BLE S ......... .... ........ .... .... ...... ....................... .... vii
LIST OF FIGURES ......... ............................... ........ ............ ix
A B STR A C T ................................................. ..................................... .. x
1 G EN ER A L IN TR O D U C TIO N ......................................................... .....................1
2 ESTIMATES OF HERITABILITY FOR EGG SIZE IN THE SERPULID
POLYCHAETE Hydroides elegans ........................................ ......... ............... 8
Introduction ............... ....... ..................................................... ................. 8
M material and M methods ......... .. .. ......... .. .. .. ............................ ....11
Collection and Spaw ning........................ .. ................................ ............... 11
H alf Sib B reeding D esign............................. ......................... ............... 14
S election n ........................................... .............................................. 16
R esu lts ......... ...... ................................................................................. 17
D isc u ssio n ........................................... ................................................................. 1 8
3 DIRECT AND INDIRECT RESULTS OF ARTIFICIAL SELECTION FOR
INCREASED EGG SIZE IN THE SERPULID POLYCHAETE Hydroides
e le g a n s .................................................................................................................. 2 4
Intro du action .................................................................................................... 2 4
M eth o d s ................................................................................................... 2 9
Collection, Spawning, and Selection .......................................29
L arval Size......................................................... 3 1
Juvenile Tube Length ............................................... ............... 32
Total Egg Energy ...... ...... ............. .................32
F ecu n d ity ....................................................... 3 4
A dult D ry W eight .............. ......................... ... ..........................................35
D ata A n a ly sis ................................................................................................. 3 5
Results .............. .......................... ...................................... 36
D direct R response to Selection .................................... ................. 36
C orrelated R responses to Selection ........................................... .....................36
Spontaneous Spawning and Sex Ratios.................................. ...............38
D iscu ssio n ................................ ............................................................................. 3 9
More + Bigger = Better... or Does It?....................... ....... ...............39
Conclusions and Future Directions ........................................ ............... 44
4 CON CLU SION S .................................. .. .......... .. .............49
L IST O F R E FE R E N C E S .......... .................... .......................................... .........................53
B IO G R A PH IC A L SK E TCH ..................................................................... ..................67
LIST OF TABLES
2-1 Number of females used to establish six individual lines in Generation 7 .............21
2-2 Half-sib breeding design; Fully nested ANOVA and variance component
analysis for egg diameter in Hydroides elegans........ ......... .................................22
3-1 Egg diameter ([tm) in replicate selected vs. control lines of Hydroides elegans
after 4 generations of selection ......... ............................................. ..... ......... 46
3-2 Total egg energy in replicate selected vs. control lines of Hydroides elegans
(calculated as |tg C egg )......... ............................................ .......... ............. 46
3-3 Absolute fecundity in replicate selected vs. control lines ofHydroides elegans
(eg g s fem ale-1).................................................. ................ 4 7
3-4 Relative fecundity based on dry weight of the female in replicate selected vs.
control lines of Hydroides elegans (eggs mg ) ...................................... ........... 47
3-5 Adult female dry weight (mg) in replicate selected vs. control lines of Hydroides
elegans ........... ........................................ ....................... .. 47
3-6 Larval volume ([tL) at 18 to 20 h post-fertilization in replicate selected vs.
control lines of Hydroides elegans.................................................... 47
3-7 Volume ([tL) of competent (5 d) larvae in replicate selected vs. control lines of
Hydroides elegans ......... ..... ........... .... .................... .. ...... 47
3-8 Juvenile tube length (mm) at 21 d in replicate selected vs. control lines of
Hydroides elegans ......... ..... ........... .... .................... .. ...... 47
3-9 Summary of direct and correlated responses to selection for increased egg
diameter in replicate selected v controls lines of Hydroides elegans ......................48
3-10 Effect of selection (treatment) on the relative numbers of males and females in
Generations 7 to 11 ................... ................... ................. .. 48
3-11 Mean percent females present at spawning (6 weeks post-fertilization) in
replicate selected vs. control lines of Hydroides elegans.............................48
4-1 Phenotypic correlations between the trait of egg size and a suite of life-history
traits in replicate selected and control lines in the polychaete Hydroides elegans ..52
LIST OF FIGURES
2-1 Response to direct selection on egg diameter for four generations of in the
polychaete Hydroides elegans ......... .............................................. ............... 22
2-2 Mean egg diameter (+ SE) before and after direct selection for increased egg
diameter in replicate selected (1,3,5) and control (2,4,6) lines of Hydroides
elegans ........... ........................................ ........................ 23
3-1 Cumulative response to selection for increased egg diameter in H. elegans was
positive through G generation 11 ........................................ .......................... 46
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
LIFE-HISTORY CONSEQUENCES OF ARTIFICIAL SELECTION FOR
INCREASED EGG SIZE IN Hydroides elegans (POLYCHAETA: SERPULIDAE)
Cecelia M. Miles
Chair: Marta L. Wayne
Goals of this study were to estimate narrow-sense heritability (h2) for egg size in
the polychaete worm Hydroides elegans, to examine direct and correlated responses to
selection on increased egg size in a suite of life-history characters, and to explore how
these correlations changed as egg size was increased by artificial selection.
Narrow-sense heritability is a predictor of short-term response to selection. Many
quantitative models have invoked selective response in egg size as a key transitional
element in the evolution of life histories in marine invertebrates, assuming that egg size
can and does respond to selection. I tested this assumption using a half-sib breeding
design and obtained an estimate for h2 of 0.45 for egg size in a Pearl Harbor, HI
population. I performed four generations of artificial selection for increased egg size that
allowed me to estimate cumulative realized heritability in the same population as 0.58.
Artificial selection resulted in a direct response of 2.5 op in egg size relative to the
common base population. Though I predicted negative correlated responses in some
traits under selection, none was measured. A positive correlated response to selection
was observed in fecundity, total egg energy, and larval size at competence, relative to
control lines. No significant correlated response was observed in early larval size,
juvenile tube length, or adult dry weight. H. elegans is a protandrous hermaphrodite, and
sex ratio data across generations indicated that large-egg selected lines were changing sex
earlier than control lines. I propose that this earlier sex change resulted in selected
individuals spending less time as males relative to controls, consequently decreasing their
Phenotypic correlations between egg diameter and the traits outlined above were
estimated, and control line means showed no significant correlation with egg size, with
the exception of the positive correlation for larval size at competence. Unexpectedly,
phenotypic correlations among selected line means were significantly negative for
fecundity, but positive for female dry weight. The positive correlation for larval size at
competence also persisted after selection. This may indicate where further work could be
useful in examining trade-offs between egg size and other life-history traits.
The diversity of larval form and function among marine invertebrates has
fascinated researchers for decades. Thorson (1950) was among the first to recognize
broad-scale patterns in developmental mode and their strong correlation with specific
life-history traits (egg size and fecundity). A useful classification of larval types has been
based on this observation that numerous marine invertebrates produce high numbers of
relatively small, energy-poor eggs, while many others produce fewer, larger, energy-rich
eggs. A closely associated dichotomy involves the coexistence of planktotrophic and
lecithotrophic larval development. Many marine invertebrates produce pelagic, feeding
(planktotrophic) larvae, and these are generally hatched in large numbers from small
eggs. Other, often closely related, species produce lecithotrophic larvae that do not
depend on exogenous food in order to reach metamorphosis, and these hatch in smaller
numbers from larger eggs. Phylogenetic analysis suggests that the transition between
these life-history modes has occurred many times (Wray 1996, Duda and Palumbi 1997,
Cunningham 1999, Rouse 2000, McEdward and Miner 2001).
The adaptive significance of developmental mode has received considerable
attention. Selection pressures such as predation, starvation, and dispersal have been
proposed for developmental mode evolution (Thorson 1950, Chia 1974, Strathmann
1985), and increases in egg size are often cited as important transitional states between
modes (Wray and Raff 1991, Hart 1996, Miner et al. 2005). The suite of characteristics
that defines a larval type has not evolved in isolation from post-larval and adult
characteristics. Egg size and fecundity are two members of a much larger complex of
life-history traits that co-evolved, presumably, under the influence of natural selection
(Havenhand 1995, Ramirez Llodra 2002). Egg size and fecundity are both directly
related to parental fitness (Roff 1992, Bernardo 1996) and many authors have mentioned
the importance of interactions between larval characteristics and other parts of the life
cycle (e.g., Vance 1973a,b, Christiansen and Fenchel 1979, Todd and Doyle 1981,
Strathmann 1985, Moran 1994, Pechenik et al. 1998, Gimenez et al. 2004). Few efforts
have been made to examine the implications over the entire life cycle of selective
pressures on these important and inter-related fitness components.
Many studies of marine ecology have used a comparative approach to assess the
influence of one life-history stage on another. For instance, maternal environmental
conditions influence larval performance in various groups (e.g., Laughlin and French
1989 for the crab Rhithropanopeus harrisii, George 1999 for the seastar Pisaster
ochraceus, Qiu and Qian 1997 for the polychaete Hydroides elegans, Gimenez and Anger
2003 for the crab Chasmagnathus granulata). Studies have demonstrated that larval
experience is linked to juvenile quality (e.g., Jarrett and Pechenik 1997 for the barnacle
Semibalanus balanoides, Wendt 1998 for the bryozoan Bugula neritina, Phillips 2002 for
the bivalve mollusk Mytilus galloprovincialis, reviewed in Pechenik et al. 1998). Size at
hatching affects growth, time to maturity and survivorship in the gastropod Nucella
ostrina (Moran and Emlet 2001). Size at settlement influences survival in the ascidian
Botrylloides violaceus (Marshall et al. 2006). Hilbish et al. (1993) discussed the common
assumption that variation in larval and juvenile growth rates have a common genetic
basis, and found that significant genetic variation for larval shell growth in the bivalve
Mercenaria mercenaria is not related to the genetic variation in shell growth during the
juvenile period. Pechenik et al. (1996) found no correlation of growth rates between
larval and juvenile stages in two species of the mollusk Crepidula. Delayed
metamorphosis is associated with reduced juvenile or adult performance most often in
non-feeding larval forms (e.g., Wendt 1996 for the bryozoan Bugula neritina, Pechenik
and Cerulli 1991 for the polychaete Capitella sp. I, Pechenik et al. 1993 for the barnacle
Balanus amphitrite, and Maldonado and Young 1999 for the sponge Sigmadocia
caerulea). Delaying metamorphosis in the feeding larvae of polychaete Hydroides
elegans negatively affects juvenile survival and growth rates whether or not they are fed
during the delay (Qian and Pechenik 1998). Delayed metamorphosis in the feeding
larvae of the sipunculan Apionsoma misakianum results in reduced juvenile growth
whether or not larvae are fed, but does not affect juvenile survival (Pechenik and Rice
Empirical studies have examined the relationship between egg size and larval
characters such as rate of development and larval size at metamorphosis in a range of
taxa. For example, in opisthobranch mollusks, Havenhand (1993) reports a negative
relationship between egg size and development time. Gimenez (2002) reports that the
crustacean Chasmagnathus granulata hatching from smaller eggs takes longer to
develop, as does the seastar Pisaster ochraceus (George 1999). This negative
relationship between egg size and development time is also reported for echinoids and
asteroids combined (Emlet et al. 1987). Levitan (2000) further refined this relationship
by showing that it is negative curvilinear using data from a large group of echinoids. For
echinoids and asteroids combined the correlation between size of egg and the size of
juvenile at metamorphosis is positive overall (Emlet et al. 1987), but Levitan (2000)
shows these two traits to be independent of one another in echinoids, after controlling for
Many fewer studies have used an experimental approach to directly manipulate egg
size. Sinervo and McEdward (1988) used blastomere separation to show the relationship
between reduced parental investment and resulting larval characters in two species of the
echinoid S.'i/t ii'o. l, etllf ,itn They found that larvae that develop from experimentally
reduced zygotes are smaller and take longer to reach metamorphosis than those from
whole eggs. Hart (1995) used the same technique and found reduced feeding rates in
larvae developing from experimentally reduced zygotes of S. droebachiensis. Emlet and
Hoegh-Guldberg (1997) experimentally reduced the lipid content of early zygotes in
another urchin, Heliocidaris erythrogramma, and showed a link between reduced
parental investment and juvenile performance.
Another approach to evaluating the influence of egg size on life history is to
examine a single species that shows both developmental modes (poecilogony). Levin
et al. (1987) examined the demographic consequences of divergent life-history patterns in
a single species of polychaete, Streblospio benedicti, with both planktotrophic and
lecithotrophic developmental types. They found that both types achieve similar
population growth rates.
This overview represents a sample of the diversity of taxa, traits, and methods that
have been employed to demonstrate to what extent one stage of life can influence
another. Thus, while a large and diverse body of empirical, comparative, and
experimental work has demonstrated links between one life-history stage and another, a
unified examination of egg, larval, juvenile, and adult traits, and how these correlations
influence fitness over the entire life cycle has not been undertaken. As shifts in egg size
are often invoked as important transitional states between developmental modes (Wray
and Raff 1991, Hart 1996, Miner et al. 2005), and egg size plays a critical role in many of
the quantitative modeling approaches that have been used to try and identify selection
pressures and processes that influence developmental mode evolution (Vance 1973a,b,
Smith and Fretwell 1974, Christiansen and Fenchel 1979, Perron and Carrier 1981,
Roughgarden 1989, Podolsky and Strathmann 1996, McEdward 1997, Levitan 2000,
Hendry et al. 2001, Luttikhuizen et al. 2004), I focused this study on egg size.
Though it is often reported as a fixed parameter, egg size within species displays a
large amount of variation (Jaeckle 1995, Hadfield and Strathmann 1996). Considering
marine invertebrates, intraspecific variation in mean egg size has been reported for the
barnacle Balanus balanoides (Barnes and Barnes 1965). Turner and Lawrence (1979)
found significant differences in egg size within species when examining 11 echinoderms,
and Lessios (1987) found this same result in a comparison of 13 echinoid species
sampled across the Panamanian isthmus. The presence of significant variation in egg size
has been reported within species for the starfishes Pteraster tessalatus (McEdward and
Coulter 1987), and Solaster stimpsoni (McEdward and Carson 1987), for the polar
shrimps Chorismus antarcticus and Notocrangon antarcticus (Clarke 1993), for the dorid
nudibranch Adalaria proxima (Jones et al. 1996), for the ascidian Pyura stolonifera
(Marshall et al. 2002), and for the euphausiid Thysanoessa raschii (Timofeev 2004).
Within and between-species variation in egg size was reviewed for serpulimorph
polychaetes by Kupriyanova (2001).
Certainly there is abundant phenotypic variation in egg size, but only that portion
that has a genetic basis is accessible to selection in order to influence life-history
evolution. In order for a trait to respond to natural selection it must have heritable
variation, and that variability must have fitness consequences. What component of
phenotypic variation in egg size is available to natural selection; what is the heritability of
this trait? There is evidence of negative phenotypic trade-offs between fecundity and egg
size in unmanipulated populations of a variety of species (Roff 1992). Examples are
numerous among arthropods and include crustaceans, lepidopterans, dipterans, and others
(Fox and Czesak 2000). The same trade-off has been recorded for treefrogs (Lips 2001);
and Sinervo (1990) demonstrated a negative trade-off in phenotypically manipulated
But, there is a limitation to conclusions about evolutionary response to natural
selection that can be drawn from phenotypic correlations. If the observed phenotypic
trade-off does not represent some underlying genetic antagonism among traits, then it is
not revealing the set of options available for selection on these traits. One way to detect
this relationship is by using artificial selection experiments that can reveal short-term
evolutionary responses for the selected trait. Using both the selection and the
manipulation approaches, Schwarzkopf et al. (1999) found no negative genetic
correlation between egg size and fecundity in Drosophila melanogaster. Azevedo et al.
(1996) also found no evidence for a trade-off between egg size and fecundity in their
laboratory-selected lines ofD. melanogaster. Blanckenhorn and Heyland (2004) found
no evidence of a genetic tradeoff between egg size and number in the yellow dung fly
,N tth//ll ihgll/ stercoraria. But, in another arthropod, the cladoceran water flea Daphnia,
a genetic basis for this trade-off was shown (Lynch 1984, Ebert 1993). Levin et al.
(1991) also confirmed a negative genetic correlation between fecundity and egg size in
the poecilogonous polychaete Streblospio benedicti, supporting the assumption of
trade-offs between these traits.
Goals of this study were to investigate the evolvability of the trait of egg size, to
examine the direct and correlated responses to selection on increased egg size in a
specified set of life-history characters that span the life cycle, and to explore how these
correlations change as egg size is increased by artificial selection in the polychaete worm
Hydroides elegans Haswell, 1883. Using adult H. elegans collected from the wild,
spawned in the laboratory, and raised under common environmental conditions, variation
in egg size was documented and heritability for this trait was estimated using a sib
analysis (Falconer and Mackay 1996). Using artificial selection for increased egg
diameter, realized heritability was estimated (Hill 1972) and the results of selection were
documented in a suite of life-history characters as they are reflected through the larval,
juvenile, and adult stages. Life-history traits examined included egg diameter, fecundity,
and total egg energy (at 6 weeks post-fertilization), larval size at 18 to 20 h
post-fertilization, larval size at competence (5 d), tube length at 21 d, and adult dry
weight (at 6 weeks post-fertilization).
ESTIMATES OF HERITABILITY FOR EGG SIZE IN THE SERPULID
POLYCHAETE Hydroides elegans
Life histories represent compromises between selection on individual fitness
components and constraints acting on the organism as an integrated whole. Steams
(1992) suggested that the analysis of life-history strategies must be based on, and
consistent with, the evolution of the individual life-history traits. Few individual traits
are more fundamental to life history than resource allocation to the egg. Egg size is
directly related to parental fitness, and can also have substantial fitness effects for
progeny (Roff 1992).
Life histories in marine invertebrates are very diverse, and the adaptive significance
of developmental mode has received considerable attention. One common feature in the
life histories of many invertebrates is an obligate feeding (planktotrophic) larval stage.
Paradoxically, it can also often be observed within closely related groups that some
members lack this dependence on exogenous food and exhibit lecithotrophic
development. It has been hypothesized that nonfeeding larvae evolved as a response to
either direct or indirect selection for increased egg size (Jagersten 1972, Strathmann
1978, 1985, Raff 1987, Hart 1996). Selection on egg size has also been linked to
developmental mode (Hadfield and Miller 1987), larval development time (Sinervo and
McEdward 1988), larval form (Strathmann 2000), size at settlement (Strathmann 1985),
survivorship (Bridges and Heppell 1996), and fertilization success (Levitan 2000).
Vance's (1973a,b) fecundity-time model regarding the evolution of life histories in
marine invertebrates began a series of quantitative modeling efforts focused on egg size
as a key trait (Christiansen and Fenchel 1979, Caswell 1981, Perron and Carrier 1981,
Grant 1983, Roughgarden 1989, McEdward 1997, Levitan 2000, Luttikhuizen et al.
2004). While the quantitative modeling efforts regarding evolution of life histories in
marine invertebrates have become increasingly refined, there has been little research into
the genetic architecture of the important trait of egg size in marine invertebrates.
In order for a trait to respond to natural selection it must have heritable variation,
and that variability must have fitness consequences. Numerous studies have found that
egg size typically exhibits phenotypic variability across a wide range of taxa (Hadfield
and Strathmann 1996). Phenotypic variation (Vp) has several components including the
genetic (VG), the environmental (VE), and the interaction between them (VGXE). Unless
phenotypic variation represents some underlying genetic variation, it is not heritable and
is consequently unavailable to natural (or artificial) selection. VG can be further
partitioned; the component that accounts for the resemblance between offspring and
parents in sexually reproducing organisms is called additive genetic variance (VA). The
proportion of Vp that is made up of VA is a quantity called narrow-sense heritability
(h2 = VA Vp). This value is a measure of the resemblance between parents and offspring
after sexual recombination and, therefore, is a predictor of short-term response to
selection (Falconer and Mackay 1996).
Among protostome invertebrates, most of the investigations into h2 of egg size have
been focused on the Arthropoda (reviewed by Fox and Czesak 2000): especially the
Insecta, a mostly terrestrial group that includes Diptera (flies and mosquitoes),
Lepidoptera (moths and butterflies), and Coleoptera (beetles), among others. Many
experiments examined egg size in Drosophila melanogaster and demonstrated that there
is a significant genetic component to variation for this trait (Warren 1924, Bell et al.
1955, Azevedo et al. 1997, Schwarzkopf et al. 1999). Among lepidopterans, Fischer et
al. (2004) estimates h2 for egg size of ca. 0.4 in the butterfly Bicyclus anynana and
Harvey (1983) reports h2 for egg weight in the budworm Choristoneurafumiferana at
0.75. Fox (1993) estimates h2 of egg size for the seed beetle Callosobruchus maculatus
at 0.43 to 0.59 and 0.60 to 0.74 using two different designs. Czesak and Fox (2003)
calculate realized h2 for egg size in the seed beetle Stator limbatus of 0.36 to 0.55. Thus,
there is abundant evidence of VA for egg size in insects, and estimates of h2 are generally
moderate to large.
The only marine invertebrate to receive similar attention was the poecilogonous
polychaete Streblospio benedicti, where a reciprocal mating design used to examine the
genetic components of life-history traits gave a heritability estimate of 0.75 for egg
diameter (Levin et al. 1991). With the exception of this one study on a species with an
unusual life history that includes producing both planktotrophic and lecithotrophic larvae,
empirical support for theoretical assumptions concerning the heritability of egg size in
marine invertebrates has not been examined.
The serpulid polychaete Hydroides elegans Haswell, 1883 is an excellent choice as
a model organism for estimating h2 for egg size in a marine invertebrate for several
reasons. It produces a large number of planktotrophic larvae that can be cultured in the
laboratory with relative ease. The larvae will settle and metamorphose in culture in the
presence of natural biofilms (Carpizio-Ituarte and Hadfield 1998). The generation time is
reasonably short, 16 to 28 days at optimal temperature, salinity, and food concentrations
(Qiu and Qian 1998). Both larvae and adult worms can be fed the chrysophyte alga
Isochrysis galbana, which is easily cultured in the laboratory. H. elegans develops in the
water column with no maternal care.
The objective of this study was to obtain estimates of narrow sense heritability for
the trait of egg size in Hydroides elegans. I calculated h2 using three different
techniques. First, I constructed a half-sib breeding design and used a nested ANOVA for
analysis (Falconer and Mackay 1996). Next, Restricted Maximum Likelihood Analysis
(Quercus Quantitative Genetics Software, University of Minnesota, USA) was performed
on the same data set. Last, artificial selection on egg size was performed and cumulative
realized heritability was calculated according to the methods of Hill (1972).
Material and Methods
Collection and Spawning
This project was completed at Kewalo Marine Laboratory, Honolulu, Hawaii using
protocols developed there by various workers. Wild worms were collected in November
2002 from VexarTM screens suspended for 1 month from a floating dock at Ford Island,
Pearl Harbor, Hawaii. The sex of each worm was determined and gametes secured by
agitation of the calcareous tube surrounding each adult, resulting in release of gametes
(Unabia and Hadfield, 1999). Gametes from 270 female worms were used to establish
the initial laboratory population. At least 120 males were used, but the exact number of
males was not determined.
I used a Nikon CoolpixTM 990 camera mounted on an Olympus compound light
microscope to obtain digital images of eggs from each female immediately after their
release. Mean egg diameter for each mother (n = 10 eggs mother-) was obtained from
these images using NIH Image software. Larval cultures were established at densities of
5 tolO larvae mL-1 in 1L plastic tripour beakers and raised in pooled cultures (4 to 5
mothers and 2 to 3 fathers beaker-) for Generations 1 through 3. Each beaker had
different and unique parents. Cultures were maintained at 250C, and fed 6 X 104
cells mL-1 day-'of the chrysophyte alga Isochrysis galbana (Tahitian strain). Isochrysis
galbana was cultured in the lab at room temperature and constant light, using f/2 media
(Guillard 1975) and used during exponential growth phase. All embryos were allowed to
hatch in FSW (0.22 |tm filtered seawater) and swimming larvae were counted
approximately 12 h post-fertiization. At that time larval cultures were established at
maximum densities of 10 larvae mL-1 and larvae were fed for the first time. Larvae were
fed on Day 2 by adding the appropriate volume of algae to the larval culture beaker
without changing water or beakers. On Day 3 and Day 4 larvae were fed and both water
and beakers were changed. Day 5 larvae were considered competent to settle and
Small transparent plastic chips (2.5 X 1.5 X 0.08 cm; K&S Engineering, Chicago
IL, #1306) were placed in flow-through seawater tables for 10 days in order to
accumulate natural biofilm, which has been shown to induce settlement in H. elegans
(Hadfield et al. 1994). One biofilmed chip was placed in each well of a standard ice cube
tray along with 15 mL FSW, I. galbana (6 X 104 cells mL-1), and 20 to 25 competent
Larvae and newly settled juveniles were fed daily at this food density without
changing the water for 5 days. This procedure allowed a wide window of opportunity for
settlement and avoided inadvertent selection for early settling individuals. Ice trays were
kept at room temperature (-240C) in a PlexiglasTM rack on the bench top with ambient
illumination. The position of trays within the rack was randomized daily. On Day 10
(post-fertilization), each chip was removed from the tray and all animals but one were
removed. The survivor was determined by position on the chip, with that individual
closest to the center selected to be the survivor. Thus, no inadvertent selection for early
or late settling individuals, or for body size, took place during this process. At this point,
the food level was increased 6-fold (36 X 104 cells mL1) and worms were fed with the
daily water change (FSW) until animals had reached 6 weeks of age. At maturity (here
defined as 6 weeks), individual worms were placed in Petri dishes and spawned as before.
H. elegans is a protandrous hermaphrodite. Since I was measuring a maternal trait (egg
size), I chose a generation time of 6 weeks in order to assure that 50 to 60% of the
spawning population would be female. Three generations were so raised under
laboratory conditions to minimize variance due to maternal effects. Two hundred or
more females were spawned to produce each of these two subsequent generations (217 in
Generation 2 and 236 in Generation 3).
Within Generation 3 a subset of individual crosses was established to setup the
breeding design experiment described below. Alongside the pooled cultures of mixed
parentage described above, individual larval cultures consisting only of offspring from a
unique mother crossed with a known father were raised under conditions identical to
those described above.
Half Sib Breeding Design
A subset of the worms in Generation 3 was used as parents for the half-sib design
experiment (Falconer and Mackay 1996) that was analyzed in Generation 4. Ten males
were arbitrarily selected and their sperm was split between two females each (3 females
in one case). The result of this mating scheme is to produce offspring related in two
different ways, that is, full-siblings (within females, or dams) and half-siblings (within
males, or sires). The advantage of this design is to allow the partitioning of the
phenotypic variance according to its source (the progeny of different males (2 SIRE), the
progeny of different females mated to the same male (C2DAM), and individual offspring of
the same female withinHIN. The component of total variance attributable to sires is
1/4 VA, therefore the most straightforward estimate of h2 = 4C2sIRE/ C TOTAL (Falconer and
Parents and offspring involved in this breeding design were raised in individual and
not pooled cultures, but otherwise all conditions were as described above. A fully nested
Random Model ANOVA was then used to partition variance in egg diameter according to
its source. Type III sums of squares were calculated using the general linear model of
SAS Version 6.12 (PROC GLM, with RANDOM statement and TEST option).
The variance component data were also used to calculate the additive genetic and
residual coefficients of variation (CVA = 100 VA IX and CVR = 100 V- VA IX, where
X = trait mean) as recommended by Houle (1992). Evolvability is defined as the ability
of a trait to respond to selection (Houle 1992) and, using the assumption of directional
selection, can be calculated as IA = VA/X2.
I also used Restricted Maximum Likelihood (REML) to analyze these same data,
using the nf3 program in Quercus of R.G. Shaw and F.H. Shaw (University of Minnesota,
USA; downloaded November, 2004; www.cbs.umn.edu/eeb/events/quercus.shtml). This
program estimates the likelihood of observing the data given a set of parameters.
Iterative methods are then used to find the set of parameters that maximize this likelihood
(Shaw 1987). I used the program to perform REML on a two-generation pedigree with
mean egg diameter of each Generation 4 daughter as the single character. The sources of
total phenotypic variance were analyzed using the model Vp = VA + VD + VE. In this
model the genetic variance mentioned above (VG) is partitioned into two components, VA
(additive) and VD (dominance). Log likelihood ratio tests were used to compare the fit of
the model for significant differences (2 21) as successive variance components were
constrained to zero. Technically, this violates the assumption of the likelihood ratio test
because the null value of the parameter (i.e., variance = 0) lies on the boundary of the
feasible range. However, the test is believed to be conservative under these conditions
(Pinheiro and Bates 2000).
Linear regression of mid-offspring egg diameters onto maternal egg diameters was
not used because, while each daughter had an independent mother, they were not
independent with respect to the father. Therefore, the assumption of independence
necessary for the linear regression could not be met.
The family based design described here for Generation 4 was repeated in
Generations 5 and 6 in an attempt to increase sample size.
In Generation 7 the common laboratory population was split into six lines. Three
were established as selected lines and three as control lines. Between 12 and 16 females
were used to establish each line (Table 2-1). In subsequent generations (Generations 8 to
10), 16 mothers were kept in each line. Within selected lines, the eggs from a specific
mother were sub-sampled as described above. Measurements were made immediately
and eggs were fertilized only if the mean diameter of the eggs was greater than one
phenotypic standard deviation (op) over the Generation 6 population mean. In
subsequent generations, selection proceeded in the same manner, with mothers remaining
in the line only if the mean diameter of their eggs exceeded 1 op over the mean of the
selected mothers from the previous generation. If the mothers meeting this criterion
happened to include multiple sisters I retained no more than two sisters in a line within a
generation. The same rule, that no more than two sisters remain in a line, was also
followed for control lines. In control lines, every third or fourth female was fertilized
without regard to the size of her eggs. Otherwise, eggs were handled exactly the same as
for selected lines. All crosses were between one male and one female with no shared
parentage. Males were chosen haphazardly with respect to the mean egg size of the
clutch from which they came. Care was taken not to use siblings of either selected or
control females as fathers. Larval culture and settlement proceeded as before. Selection
continued for four generations (7 through 10).
Generally, realized heritability is the change in mean trait value over each
generation of selection (offspring mean parental population mean) divided by the
selection differential (trait mean for selected parents parental population mean). As
selection was carried out in one direction only, and the selected and control lines came
from the same base population, the realized heritability was calculated according to the
specific methods described by Hill (1972). Cumulative realized heritability was
calculated as the difference between selected and control values at each generation (X,)
multiplied by the cumulative selection differential for that generation (S,), summed over
generations (YXYS,). This quantity is divided by the sum over the square of the
cumulative selection differential for each generation (yS,2).
Expected mean squares and variance components for the half-sib breeding design
in Generation 4 are listed in Table 2-2. Narrow-sense heritability calculated using the
sire component of variance (P = 0.069) was h2 = 4C2sIRE/ 2TOTAL =0.45. Confidence
intervals were calculated based on the F-distribution, according to Knapp (1986), as
recommended for small sample sizes (Hohls 1998). The 95% confidence intervals were
large (-0.36 to 0.90) as is often the case for CI for heritability from a breeding design
(Koots and Gibson 1996, Markow and Clarke 1997). The mean egg size of Generation 4
daughters used in the nested ANOVA, grouped by mother, was 45.68 ptm
(+ 0.21 |tm SE), CVA = 3.22 and CVR = 3.59. Evolvability for the trait of egg size for the
Pearl Harbor population ofH. elegans was calculated at I = 10.35 X 10-4. Results from
the breeding design experiments in Generations 5 and 6 did not result in an increase in
sample size and in both cases the sire component of variance was not significant.
Results from the REML analysis on the Generation 4 data were not significant
using likelihood ratio tests. The full model (VA, VD, VE unconstrained) gave a negative
estimate for VD. Since this is not biologically meaningful, VD was constrained to zero.
This produced a result not significantly different than the unconstrained model
(2i = 0.735; P = 0.45). The estimate ofh2 based on this model was VA / (VA +
VE) = 0.38. However, when VA was constrained to zero the log likelihood ratios were still
not significantly different (,i = 2.519; P = 0.13).
Mean egg diameter ([tm + SE) before selection (Generation 6) was 44.78 (0.15),
and after selection the mean of the three selected lines was 49.23 (0.35). Direct selection
on egg diameter for four generations produced a shift of 4.45 |tm or 2.5 op from the
common base population mean (Figure 2-1). Figure 2-2 illustrates the mean egg diameter
([tm SE) in the base population before selection and among lines after selection began.
Egg diameter data for selected Line 1 in Generation 7 were lost due to a corrupted
computer file and therefore were not included in calculations of realized heritability.
Cumulative realized heritability was calculated using the method of Hill (1972) as 0.58.
Bootstrapping based on re-sampling with replacement, and stratified by generation and
line, was used to calculate 95% confidence intervals around this estimate. Confidence
intervals based on 1000 replicates were 0.51 to 0.66.
There has been considerable interest in the evolution of life histories in marine
invertebrates, and egg size has played a critical role in many of the quantitative modeling
approaches that have been used to try to identify selection pressures and processes that
influence the evolution of developmental mode in this group. In order for evolution to
occur, some component of phenotypic variation in egg size must be additive genetic
variation. My objective was to determine if the assumption of many models, that egg size
can and does respond to selection, is reasonable by estimating h2 for a marine
invertebrate with a wide geographic range, small egg, and obligate planktotrophic larvae.
All estimates of heritability carry the constraint that they only apply to the population
measured under those environmental conditions and are not specifically applicable to
other populations (Falconer and Mackay 1996).
I have shown that there is significant additive genetic variation for egg size in the
Pearl Harbor population ofH. elegans using various analytical techniques. The response
of 2.5 op to direct selection on egg size indicates that there is substantial potential for egg
size to respond to varying selective pressures, at least in the direction of an increase. It is
apparent in Figure 2-2 that variation within selected lines was not reduced relative to
either values before selection, or to control lines.
In a half-sib breeding design, offspring have one parent in common and the other
parent different. The degree of resemblance among half-sibs (the covariance of half-sibs)
represents half the variance of the common parent or a quarter of the additive genetic
variance (Falconer and Mackay 1996). The estimate of heritability reported here (0.45)
was based on the sire component of variance. One of the advantages of using the sire
component for this estimate is that it does not include maternal or dominance effects
(Falconer and Mackay 1996). It is possible to use this breeding design to estimate these
effects by examining the variance within dams (C2DAM), since it includes both maternal
(common environment) and dominance effects, and comparing it to the estimate based on
the sire component. However, since my estimate of G2DAM was not statistically
significant (P = 0.667), and the proportion of Vp explained by this estimate was
essentially zero (Table 2-2), further extrapolation based on this estimate cannot provide
useful information. Maternal effects are widely recognized as important components of
Vp (Mousseau and Fox 1998), although this is more common in mammals and animals
with longer periods of maternal care. Hydroides elegans has external fertilization. All
development up to a competent larva takes place in the water column and there is no
evidence of parental care. However, it is possible that the small sample size in my
breeding design (10 sires, 21 dams) resulted in an overestimate of o2SIRE, obscuring my
ability to detect C2DAM and estimate maternal effects. Therefore, based on these data, I
cannot say whether maternal effects are contributing significantly to the total phenotypic
Cumulative realized heritability is considered the most precise of the various
methods for determining narrow-sense heritability (Hill 1971). The cumulative realized
heritability estimate does include maternal effects, since it is based on response to
selection, but passing six generations under laboratory conditions prior to the beginning
of selection minimized the phenotypic expression of any maternal effects. The half-sib
breeding design gave only a slightly lower estimate of h2 (0.45) compared to the
cumulative realized heritability (0.58). The fact that two estimates are similar lends some
support to the idea that maternal and dominance effects are not large in this population
with respect to this trait.
It is thought that additive genetic variation for traits closely related to fitness, such
as egg size, should be under constant directional selection, eroding VA and resulting in
low heritabilities (Gustafsson 1986, Roff and Mousseau 1987, Falconer and Mackay
1996). This trend is generally supported for life-history traits as compared to
morphological traits (Mousseau and Roff 1987). However, many exceptions can be
found (e.g., Levin et al. 1991, Gibson 1993, Fox and Csezak 2000, Edmands 2003),
including this work. Houle (1992) pointed out that because h2 is a ratio of VA to Vp, the
magnitude of Vp greatly influences the outcome: low h2 could be the result of low VA or
high Vp. Therefore, it is a biased estimate of additive genetic variance and potentially a
misleading indicator of the ability of a particular trait to respond to selection. Using
coefficients of variation, that is VA standardized by the trait mean rather than Vp, is
recommended as a more useful approach. My estimates of CVA and CVR give very
similar values (3.22 and 3.59 respectively) implying that nearly half of the total variation
present is additive genetic. This is consistent with my estimate of h2.
Using the traditional methods of estimating h2, and also by calculating CVA and IA, I
have found significant levels of additive genetic variance for the trait of egg size in the
Pearl Harbor population of H. elegans. Several hypotheses have been offered to explain
the apparent paradox of finding VA for a life-history trait that is closely linked to fitness
and, presumably, under directional selection. One such hypothesis is the presence of
microevolutionary trade-offs. Microevolutionary trade-offs occur when a change in one
trait that acts to increase fitness is linked to a change in another trait that decreases fitness
(Stears 1992). These relationships can act to constrain the simultaneous evolution of
suites of fitness-related traits and thereby maintain additive genetic variance in a
population. Examination of the correlated responses of other fitness-related traits to
artificial selection for increased egg size is one way to elucidate this relationship.
Table 2-1. Number of females used to establish six individual lines in Generation 7
Line Number of Females
Table 2-2. Half-sib breeding design; Fully nested ANOVA and variance component
analysis for egg diameter in Hydroides elegans (number of observations 550,
s = 10, d= 2.1, k= 2.6)
Source df EMS F P Proportion
(X 10-5) Vp explained
Daughter (Dam (Sire))
* Generation 6
l Generation 11
40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55
Egg diameter ([tm)
Figure 2-1. Response to direct selection on egg diameter for four generations of in the
polychaete Hydroides elegans. Selection began from a common base
population in Generation 6 and continued from Generation 7 through 10
- Selected Line 1
--- Selected Line 3
A Selected Line 5
- -X- Controls Line 2
- -X- Controls Line 4
- -0- Controls Line 6
Gen 3 Gen 4 Gen 5 Gen 6 Gen 7 Gen 8 Gen 9 Gen 10 Gen 11
Figure 2-2. Mean egg diameter (+ SE) before and after direct selection for increased egg diameter in replicate selected (1,3,5) and
control (2,4,6) lines ofHydroides elegans
DIRECT AND INDIRECT RESULTS OF ARTIFICIAL SELECTION FOR
INCREASED EGG SIZE IN THE SERPULID POLYCHAETE Hydroides elegans
Marine invertebrates exhibit a wide diversity of developmental strategies and their
mode of development is thought to have important effects on dispersal and population
structure (e.g., Hedgecock 1982, Levin and Huggett 1990, McMillian et al. 1992, Hoskin
1997, Collin 2001), geographic range (Scheltema 1989, Kohn and Perron 1994, but see
Emlet 1995), and rates of speciation and extinction (Strathmann 1985, Jablonski 1986,
1987). Egg size has been the focus of a series of quantitative modeling efforts to
analytically assess the patterns we observe in marine invertebrate life histories.
Beginning with Vance's influential fecundity-time model (Vance 1973a,b), numerous
authors have examined trade-offs between egg size, fecundity, mortality and their
relationship with the evolution of different reproductive strategies (Christiansen and
Fenchel 1979, Caswell 1981, Perron and Carrier 1981, Grant 1983, Roughgarden 1989,
McEdward 1997, Levitan 2000). Another set of models examines fertilization kinetics,
which also can pose selective pressures on egg size and thus developmental mode
(Levitan 1993, 1996, Luttikjizen 2004).
Many authors have emphasized that the suite of characteristics that defines a larval
type has evolved in conjunction with post-larval and adult characteristics, rather than in
isolation. Egg size and fecundity are two members of a much larger complex of
life-history traits that has co-evolved, presumably under the influence of natural selection
(Havenhand 1995, Ramirez Llodra 2002). Egg size and fecundity are both directly
related to parental fitness (Roff 1992, Bernardo 1996) and many authors have mentioned
the importance of interactions between larval characteristics and other parts of the life
cycle (e.g., Vance 1973a,b, Christiansen and Fenchel 1979, Todd and Doyle 1981,
Strathmann 1985, Moran 1994, Pechenik et al. 1998, Gimenez et al. 2004). Few efforts
have been made to examine the implications over the entire life cycle of selective
pressures on these important and inter-related fitness components.
Much of life history theory is based on the assumption of trade-offs among traits
that contribute to fitness, that is, that natural selection cannot produce unlimited
simultaneous increases in individual fitness components. This could result in creating
theoretical optima based on trade-offs between fitness components (Roff 1992, Steams
1992). Fitness components, and therefore trade-offs between them, must have a genetic
basis for selection to act. Alleles that increase fitness without a cost are expected to
achieve a frequency of one very quickly; therefore genetic correlations between
fitness-related traits that have been subject to simultaneous selection are expected to
Several methods are in use to examine the genetic architecture of fitness-related
traits. Genetic correlations may be estimated by examining the resemblance between
relatives using designs similar to those for estimating heritabilities (Falconer and Mackay
1996, Lynch and Walsh 1998). These correlations are caused by pleiotropy, where traits
of interest are influenced by common genes; however, they may also be caused by
linkage disequilibrium between distinct loci, each influencing a different trait. But
genetic correlations between life-history traits can be difficult to extrapolate from
breeding designs. Correlations estimated in this manner are dependent on linkage
equilibrium and other requirements (Rose et al. 1990), and standard errors around these
estimates are often very large (Koots and Gibson 1996).
A second, and more direct method that can be used to look for evidence of
evolutionary trade-offs is artificial selection. Artificial selection experiments allow
workers to manipulate the frequency of alleles associated with a phenotypic value of a
selected character by controlling the parental genetic contribution to the next generation
in laboratory populations, thereby producing divergent phenotypes. If selected lines
differ from control lines in traits other than the selected trait (and environmental variation
has been kept constant across generations), this response indicates an additive genetic
correlation between the traits (Cheverud 1984). Repeated episodes of selection over
multiple generations are more likely to reveal potential interactions between fitness
components that may be subtle or difficult to measure across a single generation.
Artificial selection experiments are a particularly powerful way to explore trade-offs
(Reznick 1985, 2000, Brakefield 2003). However, selection experiments are subject to
experimental or procedural artifacts that may be difficult to control, and the complexity
and interactive nature of response to selection can produce varying outcomes (Harshman
and Hoffman 2000).
Selection experiments on life-history traits have a long history, with the
preponderance of work in invertebrates involving insects. For example, Englert and Bell
(1969, 1970) selected on larval development time in the flour beetle Tribolium
casteneum, Roff (1990) selected on wing dimorphism (an indicator of development time)
in the sand cricket Gryllusfirmus, Bradshaw and Holzapfel (1996) selected on
development time in the pitcher-plant mosquito Wyeomyia smithii, and Palmer and
Dingle (1986) selected on wing length (an indicator of dispersal ability) in the milkweed
bug Oncopeltusfasciatus. Focusing on correlated responses in egg size, Tucic et al.
(1998) and Seslija and Tucic (2003) found that bean weevils, A.,luilvteli Ie'\ obtectus,
selected for faster development times lay significantly larger eggs. A similar correlated
response in egg size to selection on development time is reported for Drosophila
melanogaster by Bakker (1969). Selection experiments involving life-history traits in D.
melanogaster are numerous, especially focusing on trade-off between longevity,
fecundity, and larval development time (e.g., Tantawy and El-Helw 1966, Partridge and
Fowler 1992, Zwaan et al. 1995, Nunney 1996).
Several studies have examined response to selection on egg size. A significant
response to selection toward both large and small eggs has been reported in Drosophila
melanogaster in several studies (Bell et al. 1955, Parsons 1964, Schwarzkopf et al. 1999).
Azevedo et al. (1997) found evidence of a selective advantage to larger eggs in D.
melanogaster, including increased embryonic viability, hatchling weight, and larval and
pre-adult development rates. Czesak and Fox (2003) selected directly on egg size in the
seed beetle Stator limbatus and report correlated responses in lifetime fecundity and
female body mass.
Among marine invertebrates genetic correlations for selected life-history characters
have been reported using breeding designs. In mollusks, Ernande et al. (2003) report a
positive correlation between reproductive plasticity, growth and survival for the bivalve
Crassostrea gigas. Hilbish et al. (1993) discuss the common assumption that variation in
larval and juvenile growth rates have a common genetic basis, but find no significant
correlation between variation for larval and juvenile shell growth in the bivalve
Mercenaria mercenaria. Gibson (1993) explored genetic correlations among life-history
traits in the poecilogonous opisthobranch Haminoea callidegenita and found that many
reproductive traits showed significant genetic correlations while exhibiting little or no
phenotypic correlations. In crustaceans, Arcos et al. (2004) estimate the genetic basis for
a series of reproductive traits in the white shrimp Penaeus vannamei and report no
correlation between egg diameter and other traits, but did find a positive correlation
between egg number and egg total protein. Levin et al. (1991) investigated the genetic
architecture of the poecilogonous polychaete Streblospio bendicti using a diallel design
and confirmed a negative genetic correlation between egg size and fecundity, as well as
the expected positive genetic correlation between fecundity and female body size.
Ernande et al. (2004) used a nested half-sib mating design to examine five traits in
Crassostrea gigas: larval shell length, size at settlement, weight after metamorphosis,
juvenile weight, adult weight, but did not examine egg size. They report that larval
development rate is positively genetically correlated with size at settlement, but
negatively correlated with both metamorphic success and juvenile survival. Little work
has been done to examine the consequences across the life cycle of shifts in egg size.
The purpose of this work was to examine the direct and correlated results of
artificial selection to increase egg size in Hydroides elegans Haswell, 1883. H. elegans is
a marine, serpulid polychaete worm that produces numerous planktotrophic larvae and
can be cultured in the laboratory over multiple generations. Larvae will readily settle and
metamorphose in culture in the presence of natural biofilms (Carpizio-Ituarte and
Hadfield 1998). Fertilization and larval development takes place in the water column
with no maternal care and H. elegans develops as a protandrous hermaphrodite (Ranzoli
1962). The generation time is 16 to 28 days at optimal temperature, salinity, and food
concentrations (Qiu and Qian 1998). Both larvae and adult worms can be fed the
chrysophyte alga Isochrysis galbana, easily cultured in the laboratory. Using artificial
selection for increased egg diameter, correlated responses were documented in a suite of
life-history characters: egg diameter, fecundity, total egg energy (at 6 weeks
post-fertilization), larval size at 18 to 20 h post-fertilization, and at competence (5 d),
juvenile tube length at 21 d, and adult dry weight (at 6 weeks).
Collection, Spawning, and Selection
Wild Hydroides elegans were collected at Ford Island, Pearl Harbor, Hawaii in
November 2002. Details of the methods for spawning, larval culture, measurement of
egg diameter and selection are described in detail in Chapter 2. Field collected worms
were raised under controlled environmental conditions for six generations before
selection began in Generation 7. Between 12 and 16 females were used to establish each
line in Generation 7 (Table 2-1). In subsequent generations (Generations 8 to 10), 16
mothers were kept in each line. All crosses were unique and the fathers were chosen
haphazardly without regard to the mean egg size of the clutch from which they hatched.
No crosses were permitted between brothers and sisters within lines. Three replicate
control and three replicate selected lines were established with 128 worms in each line,
for a total of -768 worms in each generation. All females were spawned at 6 weeks
post-fertilization and females were retained in the selected lines if the mean diameter of
their eggs was > 1o than that of the previous generation. For control lines, every third
female spawned was retained in without regard to egg diameter.
Larval culture was performed according to methods described by Unabia and
Hadfield (1999). Cultures were maintained at 250C, and fed 6 X 104 cells mL-1 day-of
the chrysophyte alga Isochrysis galbana (Tahitian strain). Larvae were competent to
metamorphose at 5 d post-fertilization, when I exposed 20 to 25 larvae to individual
small plastic chips with natural biofilm as the settlement cue. Each chip was isolated in a
single well of a standard ice cube tray filled with 15 mL of 0.22[tm filtered seawater
(FSW) and algal food was maintained at 6 X 104 cells mL-1 day-. On Day 10 all
juveniles but the one nearest the center of the chip were killed by destroying the tube, so
that growth from that point took place at a density of 1 worm well-'. Food density was
increased 6-fold and water, food, and container (ice tray) were all changed daily until
adults were spawned.
I observed that some female worms in both selected and control lines
spontaneously released a few eggs (from 10 to a few hundred eggs) shortly after the daily
food and water change. I collected all the early-spawned eggs that were produced after
the daily food and water change over a four- day period (October 10 to 14 2003) prior to
the scheduled spawning of Generation 9 using a Pasteur pipet. There were 25 selected
and 7 control females (n = 32 females) that produced early-spawned eggs over this
four-day period. I began the scheduled spawning of these same females two days later.
Therefore, the timing of the sample of early-spawned eggs ranged 6 to 12 d prior to the
scheduled spawning date for these individuals. I used digital imagery (Chapter 2) to
measure mean egg diameter of these early-spawned eggs (n = 10 eggs female-) and
compared this to mean egg d measured at the time of induced spawning (later-spawned
eggs) of the same female.
Selection continued for four generations with lines terminated at Generation 11,
when adult dry weight, total egg energy, fecundity and final egg diameter were
determined. Larval size at two different time points, and juvenile tube length were
measured in Generation 10, as no larvae or juveniles were available after termination of
At 18 to 20 h post-fertilization sub-samples of trochophore-stage larvae were taken
from nearly all of the crosses (between 11 and 16 of the 16 crosses in each line) in
Generation 10. Any crosses for which I had not obtained video by the time they had
developed to 20 h were not sampled. Larvae from a single cross were placed on a clean
slide and excess water was removed with a micropipette. A cover slip was placed on
each slide using ScotchTM tape to elevate the slip above the slide. Using a technique
similar to that used by Leonardos and Lucas (2000) for bivalve larvae, videotapes of the
swimming larvae were obtained using a Sony Digital Video Camera Recorder (Model
DCR-TRV30) mounted on an Olympus light microscope. A stage micrometer was used
as a scale for each series of shots. Videos were downloaded to a Macintosh G4 where
iMovie software was used to grab still photos of larvae (n = 10 larvae cross-1). NIH
Image software was used to measure the width across the prototroch, and the distances
from the prototroch to the apex of the episphere and hyposphere, respectively. Larval
size was then calculated as the volume of two circular cones (V = 7 r2h/3; r = 1/2
prototroch width and h = distance to apex of episphere or hyposphere). The same process
was repeated for competent larvae at 5 d post-fertilization, including between 9 and 15 of
the 16 crosses in each line (n = 10 larvae cross-1) in Generation 10.
Juvenile Tube Length
Shortly after metamorphosis a calcareous tube is constructed by the juvenile. The
tube lengthens as the worm grows, and tube length is a reasonably reliable indicator of
juvenile size (Qian and Pechenik 1998, Qiu and Qian 1998). This indirect method of
measuring juvenile size was chosen to minimize loss of individuals from laboratory lines,
since removal from the tube results in death. Images of juvenile tubes at 21 d were
obtained using the Nikon CoolpixTM 990 described above, attached to an Olympus
dissecting microscope. Each individual chip was removed from the ice tray and placed in
a plastic Petri dish along with a 1cm length of ruler. Images were obtained for all
offspring (n = 8 juveniles cross-) from between 10 and 16 crosses from each line in
Generation 10. The images were analyzed for tube length using NIH Image software.
Total Egg Energy
In Generation 11 each sub-sample that had been used to measure egg diameter was
placed in a cryotube (FisherBrand 0.5 mL screw-top tube) and frozen at -200C. Frozen
samples from each line were later analyzed for organic content using a
micro-modification of the dichromate oxidation technique (Jarrett and Pechenik 1997,
Gosselin and Qian 1999) against a glucose standard (0 to 20 |tg C).
Eggs were thawed at room temperature and distilled water was added to raise the
volume to 200 [LL. Four replicate sub-samples of 10 pL were obtained and counted on a
Petri dish using a dissecting microscope and a hand-counter. This allowed calculation of
the total number in the sample that was needed to complete the fecundity estimate (see
below). The counted samples were removed from the Petri dish using a micropipette and
placed in a 13mm glass test tube. Between 300 and 500 eggs were placed in each of
three replicate tubes from a particular cross. Sixteen samples, each corresponding to a
different randomly selected cross from each line, were prepared. Tubes were allowed to
stand until the eggs had settled to the bottom and excess water was removed using a
The eggs were incubated in 1 mL concentrated (85.5%) phosphoric acid (15 min at
1050C) to remove excess chloride. Tubes were cooled to room temperature and 1 mL of
0.04% acid dichromate was added to each tube to oxidize the sample. Tubes were mixed
on a vortex mixer followed by incubation at 1050C for 15 min. Reduction in dichromate
concentration indicates the amount of organic carbon oxidized.
Samples were diluted by adding 5 mL distilled water followed by vortex mixing. A
0.5 mL aliquot was removed from each tube and combined with 4.5 mL of cadmium
iodide starch reagent (Parsons et al. 1984) and allowed to stand at room temperature for
20 min. Each sample was then diluted again by adding 5 mL distilled water and
examined spectrophotometrically (Kontron Instruments, Model Uvikon 930) at 575 nm.
Total energy was estimated as |tg C equivalents in glucose and converted to mJ using the
relationship 1 |tg C = 39 mJ (McEdward and Carson 1987).
Though widely used in comparative studies of egg energy in marine invertebrates,
the dichromate oxidation technique is known to underestimate energy contributed by
proteins (Gosselin and Qian 1999). Pernet and Jaeckle (2004) analyzed this technique
across species within annelids and echinoderms and conclude that the likely cause is that
different proteins oxidize to different degrees in the acid dichromate reaction. Their
comparison points out that across a wide range of species, developmental modes, and egg
sizes there is probably a wide range of differential allocation of various proteins and
because of this the absolute value of energy egg-1 may be underestimated using this
assay. Here I used the acid dichromate technique for a relative comparison of mean total
energy egg- within species, over a small range of egg sizes after short-term selection.
Data presented by Pemet and Jaeckle (2004) for planktotrophic annelids show that error
of total energy estimates based on this technique within species is generally small (e.g.,
estimates for serpulid polychaetes include Serpula columbiana with 0.65 mJ egg 1
0.015 SE, or 2.3% and Hydroides sanctaecrucis with 0.40 mJ egg 1 0.023 SE, or 5.8%)
and therefore unlikely to be important over the small changes in egg size considered here.
There are two possible ways that an increase in error could affect my results. First, in the
unlikely event that short-term selection on egg size resulted in a change in the type of
protein allocated to eggs of selected lines relative to control lines, error could be
increased. Second, if selection produced increased densities of a particular protein in
larger eggs, this could affect the error. Either of these would be an interesting outcome
and should be considered by future workers using techniques specific to these questions.
Absolute values of energy estimates based on this technique may be underestimates.
In Generation 11, after the sub-sample of eggs for measurement and total energy
oxidation were removed, the remaining eggs from each female (n = 435) were fixed in
5% formalin in FSW (-2% formaldehyde). These samples were set aside and eggs were
counted later. The samples were diluted to 30 mL using filtered seawater and a standard
50 mL graduated cylinder. The sample was mixed by inversion and 4 sub-samples of
250 p.L each were placed on a Petri dish and counted using a dissecting scope and a hand
counter. The number of eggs in the fixed sample was then calculated. The number of
eggs from the frozen samples (see above) was added to this number to obtain the absolute
number of eggs produced by each female. The relative fecundity was also calculated by
dividing the absolute number of eggs per female by her dry weight (see below) to give
Adult Dry Weight
The dry weight of each adult in Generation 11 (n = 685) was determined. Small
squares of aluminum foil were numbered and pre-tared on a Cahn (Model C-35)
micro-balance. Immediately after spawning sex of the worm was recorded, and each
individual was removed from the calcareous tube using forceps to crack the tube. Worms
were rinsed in distilled water and placed on aluminum foil squares. Specimens were
placed in a drying oven at 750C for 48 h. Worms were re-weighed on the Cahn balance
to the nearest 0.1 |tg and the weight of the foil was subtracted providing a post-spawn
weight for each individual.
I used a fully nested mixed model ANOVA with fixed effects at the level of
treatment only with lines nested within treatments and crosses nested within lines for the
traits egg diameter, total egg energy, larval sizes at 18 to 20 h and at 5 d, and juvenile
tube length (all nested at all levels between selected and control lines). Data were
visually checked for radical departures from normality and heteroscedasticity and none
were found. For the analysis of adult dry weight and fecundity the same fully nested
mixed model ANOVA was used with fixed effects at the level of treatment only and lines
nested within treatments. In order to test for an effect of selection on the relative
numbers of males and females across generations I used a model I ANOVA. In all cases
Type III sums of squares were calculated using the general linear model of SAS Version
6.12 (PROC GLM, with RANDOM statement and TEST option), c = 0.05 unless
otherwise stated. To compare mean egg d between early- and later-spawned eggs within
mothers I used a paired t-test in JMP 5.1 (SAS Institute, Inc.).
Direct Response to Selection
Mean egg diameter in the base population (Generation 6) was 44.78[tm (+ 0.15
SE); the mean egg diameter in the selected lines after selection (Generation 11) was
49.23 tm (0.35 SE). Direct selection for four generations for increased egg size resulted
in a shift of 4.45 |jm or 2.5 op from the common base population mean (Figures 2-1, 2-2).
Figure 3-1 illustrates the continued positive slope of the cumulative response to selection
through Generation 11 as the difference between the mean egg diameter of the replicate
selected and control lines. Egg diameters between selected and control lines were
significantly different at Generation 11 (P = 0.0006, Table 3-1).
Correlated Responses to Selection
Replicate lines selected for larger egg diameter were found to have a significantly
higher mean value for total energy relative to the control lines (P = 0.0350, Table 3-2).
The mean value (mJ egg'1 SE) for the selected lines was 0.37 (0.17) and for the control
lines the mean was 0.29 (0.18).
Fecundity, calculated as the absolute number of eggs produced by a female (+ SE),
was significantly higher in the large-egg selected lines relative to controls (P = 0.0304,
Table 3-3). Female worms in the large-egg lines produced a mean of 10071 (481) eggs
and females in control lines produced a mean of 7936 (418). When egg number was
standardized by the dry weight of the mother to give eggs mg-1 the difference between the
selected and control line means became less apparent, significant at a = 0.10 (P = 0.0679,
Table 3-4). Both of these analyses were also performed on datasets omitting all mothers
that had produced eggs prior to the induced spawning at 6 weeks (see Spontaneous
Spawning below) and the outcome was not changed.
No significant difference was found in the dry weight of adult females from
selected lines relative to control lines (P = 0.6567, Table 3-5). The mean dry weight
(mg SE) for females in the selected lines was 0.80 (0.019) and for controls was 0.82
(0.026). The same result was found if males and females were combined within
treatments (P = 0.5791), and for males alone (P = 0.8526). The Generation 11 mean dry
weight of adults of both sexes from the selected lines was 0.81 (0.025) and from the
control lines this value was 0.83 (0.020). Males could not be distinguished from females
on the basis of dry weight within either the control (P = 0.1974) or selected (P = 0.1698)
No significant difference was detected in larval size (volume) at 18 to 20 h between
selected and control lines (P = 0.3392, Table 3-6). The mean value for larval size
([tL SE) at 18 to 20 h for the large-egg selected lines was 1.12 X10-4 (0.064 X 10-4).
The control lines had a mean value of 1.14 X10-4 (0.047 X 10-4).
Larval size (volume) at competence was significantly larger in the selected than in
the control lines at a = 0.10 (P = 0.0894, Table 3-7). The mean size at competence in the
selected lines size ([tL SE) was 1.29 X 10-3 (0.041 X 10-3) and for the control lines it
was 1.16 X 10-3 (0.036 X 10-3).
Selected lines did not significantly differ from control lines in tube length at 21 d
(P = 0.1249, Table 3-8). Mean tube length (mm + SE) was 8.08 (0.31) in the selected
lines and 8.60 (0.10) in the control lines.
Direct and correlated responses to selection for increasing egg diameter are
summarized in Table 3-9.
Spontaneous Spawning and Sex Ratios
In an effort to explore the relationship between egg diameter and time of spawning,
I compared the mean egg diameter from spontaneously or early-spawned females to that
of eggs produced at 6 weeks by the same female (n = 32) in Generation 9. The
later-spawned eggs were significantly larger than the early-spawned within individual
mothers (paired t-test, t = 9.8582, P < 0.0001).
In an unexpected development, I observed that females in the selected lines tended
to spawn spontaneously more frequently and earlier than control females. Timing and
frequency of these events was not recorded since it was not part of the experimental
design. Within this design I controlled the absolute age of individuals at spawning (6
weeks), but not the time to sex change. At 6 weeks an individual female might have been
producing eggs for variable periods of time. Considering that within mothers
later-spawned eggs are significantly larger than earlier-spawned eggs, was it possible that
large-egg selected mothers had been producing eggs longer than control mothers? If
selected lines were changing sex earlier than control lines, one prediction would be a
higher proportion of females in selected lines relative to controls at 6 weeks. Since the
sex of each individual was recorded at the time of spawning for all generations I was able
to document an effect of selection on sex ratios, showing that the proportion of females at
6 weeks increased in the selected lines, relative to controls, across Generations 7 through
11 (P = 0.0122, Table 3-10). The mean percentage of females in selected lines ranged
from 64.70% to 74.30% and for control lines the range was 59.21% to 64.21% (Table
H. elegans has been described as a sequential hermaphrodite in all accounts since
Ranzoli (1962) clarified their status as hermaphrodites, and I also found this condition in
all control lines (as well as in thousands of individuals raised under laboratory conditions
prior to selection). However, in all 3 replicate selected lines on at least one occasion I
observed simultaneous eggs and sperm being produced by the same individual (again,
spawned at 6 weeks).
More + Bigger = Better... or Does It?
I predicted that there would be negative correlations between at least some fitness
related traits, because we expect alleles underlying advantageous genetic correlations to
be rapidly fixed in natural populations. Yet I found that large-egg selected females
produced more eggs, with higher energy content, that grew into larger larvae at
competence. Increased size at settlement is considered a selective advantage (reviewed
by Pechenik 1999). No trade-off with this apparently advantageous condition (the
production of more energy-rich eggs that grow into larger competent larvae) was
immediately evident by evaluating the initial traits of interest. This set of apparently
advantageous correlated traits raises the question what other traits offset the selective
advantage of larger egg size in natural populations?
One of the advantages of using a selection experiment to explore evolutionary
processes is that correlated responses to selection can reveal unexpected relationships
between traits that were not predicted at the start of the experiment (Harshman and
Hoffman 2000). The observation that selected worms appeared to produce eggs
spontaneously both earlier and more frequently than control lines prompted me to
examine the relative numbers of males and females across generations and I found that
direct selection for increased egg diameter produced significantly higher proportion of
females at 6 weeks of age. Considering that I also found that egg diameters are
significantly larger in females that spawn later than those that spawn earlier, I propose
that selection for increased egg size was manifest through selection for earlier sex
change. This earlier sex change resulted in selected individuals spending less time as
males relative to controls, consequently reducing the number of offspring they could
Sex allocation theory predicts that sex change (sequential hermaphroditism) is
favored when an individual experiences differential reproductive success in relation to
size (or age). According to the size-advantage hypothesis selection should favor
protandry if male reproductive success does not increase as individuals grow larger,
while female reproductive success does (Ghiselinl969). There has been broad empirical
support for this hypothesis (e.g., Charnov 1979a for the protandrous shrimp Pandalus,
Warner 1984a,b for protogynous coral reef fishes, Berglund 1990 for the protandrous
polychaete Ophryotrochapuerilis). The optimal timing of sex change within individuals
is also predicted by the model (Charnov 1979b, 1982). Sex change should theoretically
occur when an individual's reproductive value (future expectation of reproductive
success weighted by the probability of survival) as a member of its present sex declines
below the reproductive value of the opposite sex. Alterations between male versus
female investment should only be favored if the gain in reproductive value in one sex
exceeds the loss in the other sex (Leigh et al. 1976). An individual should change sex if
it can increase its lifetime reproductive value by doing so, and should change at that point
where the sum of the male and female components is maximized. The earlier sex change
that resulted as a correlated response to selection in H. elegans could reduce the
contribution of the male component to lifetime reproductive value to such an extent that
the female component cannot compensate for the loss. This assumes that males are not
able to shift their own sexual maturity earlier, or at least not sufficiently earlier to make
up for the loss in reproductive fitness resulting from terminating sperm production
earlier. Earlier sex change could occur as a result of male and female gametogenic
processes being controlled by different and unlinked genes, that is, a response that shifts
the timing of investment in oogenesis manipulates a different suite of genes entirely than
those that control the timing of spermatogenesis. It is likely that the spermatogenic
pathway has already been selected to function at the earliest possible age, and therefore
smallest size, because H. elegans is an opportunistic polychaete, and among the first
settlers into disturbed environments. Maternal biosynthesis, mobilization and
bioaccumulation of nutrients during oogenesis is an energy intensive activity and it is
reasonable to assume that there is a cost involved in pushing this activity to an earlier
time in the life cycle until it nearly, or actually, overlaps with sperm production.
Therefore, I propose that the constraint on increased egg diameter (at 6 weeks) in this
population is selective pressure to maintain the optimum male contribution to lifetime
reproduction. One question that follows on this reasoning is, if it is costly to change sex
early, then how is variation for the trait of "time to female" maintained in the population?
One possible explanation is frequency-dependent selection, that it is very profitable to be
an early sex-changing female when the trait is rare (and everyone else is male), but
becomes less profitable, even costly, as the frequency increases.
Though it is widely reported that sequential hermaphrodites change sex as their size
increases, field studies commonly report extensive overlap between male and female size
classes (Wright 1989, Sewell 1994, Collin 1995). Ranzoli (1962) surveyed field
populations ofH. elegans (as H. norvegica) over a three year period using segment
number as an indicator of size and also found considerable overlap between the sexes.
Population studies necessarily include overlapping generations; however, even with the
discrete generations raised here I observed no significant difference in somatic dry weight
between the sexes. This variation in size at 6 weeks may indicate significant genetic
variation in maximum body size, or possibly in growth rate in this population. Many of
the males that remained in both selected and control lines at 6 weeks were quite large and
the mean dry weight for males was larger (although not significantly so) in both
Given the positive correlation between egg size and egg number reported here, one
may ask how often, and under what conditions the predicted trade-off between egg size
and egg number has been empirically demonstrated. Varying environmental conditions
can result in an interaction between genotype and environment (G X E) and reveal a
trade-off. Czesak and Fox (2003) found that the magnitude of the trade-off between egg
size and fecundity varied between environments in the seed beetle Stator limbatus.
Varying food levels revealed a trade-off between increased fecundity and mortality in the
polychaete Dinophiilus gyrociliatus (Prevedelli and Zunarelli Vandini 1999, Prevedelli
and Simonini 2000). Tatar et al. (1993) nicely demonstrate the trade-off between
increased egg production and increased age-specific mortality in the beetle
Callosobruchus maculatus. A longstanding idea concerning the evolution of offspring
number is the "Lack clutch size" hypothesis in birds, that is, that clutch size has evolved
towards the size that will maximize the number of successful fledglings. Empirical tests
of the "Lack clutch", and deviations from it, continue to produce mixed results. For
example VanderWerf (1992) used meta-analysis on data from 77 different ornithological
studies and found that the hypothesis was not supported. Frequently cited examples of
the egg size-egg number trade-off in marine invertebrates are poecilogonous species such
as the polychaetes Streblospio benedicti (Levin and Creed 1986, Levin et al. 1987) and
Capitella sp. (Qian and Chia 1991) that have sympatric populations capable of producing
either planktotrophic or lecithotrophic larvae. The reproductive output of both strategies
is similar but lecithotrophs produce fewer, larger eggs while planktotrophic populations
produce smaller, more numerous eggs. The negative genetic correlation between egg size
and fecundity was confirmed for Streblospio benedicti by Levin et al. (1991). However,
poecilogony is an extremely unusual developmental strategy (Hoagland and Robertson
1988, Bouchet 1989) and may not represent a general pattern. In examining the more
typical reproductive strategies the genetic evidence of a trade-off within species
regarding egg size and egg number is often mixed or absent altogether (e.g., Czesak and
Fox 2003 for the seed beetle Stator limbatus, Schwarzkopf et al. 1999 for Drosophila
melanogaster, Allan 1984 for the copepod Mesocyclops, Stearns 1983 for the mosquito
fish Gambusia affinis, Reznick 1982 for the guppy Poecilia reticulata).
Conclusions and Future Directions
The constraint on increased egg diameter (at 6 weeks) in the Pearl Harbor
population of Hydroides elegans appears to be selective pressure to maintain the
optimum male contribution to lifetime reproduction by delaying the shift to the female
sex. Considerable genetic variation for egg size, fecundity, and the time to sex change is
present in this population and this is not surprising given that this population likely
includes substantial genetic diversity. Pearl Harbor is an extremely busy international
port, and H. elegans is a cosmopolitan member of the biofouling community that is
commonly found on ship and boat hulls, water intakes, and other underwater structures,
the population in Pearl Harbor likely has been assembled from multiple founding
One way to test the hypothesis whether the hermaphroditic life style is constraining
egg size is to perform a phylogenetic analysis that maps egg size and sex system onto a
phylogeny of family Serpulidae and reconstructing ancestral transitions. This type of
cladistic analysis was used by Rouse and Fitzhugh (1994) to test whether broadcast
spawning and planktonic larvae are pleisiomorphic for the polychaete family Sabellidae.
Another set of questions that arise from these results concerns the traits of sperm
that are produced by larger-egg selected hermaphrodites. It has been suggested (Levitan
1993, 1998) that sexual selection acts on gamete traits to produce suites of co-adapted
traits that include egg size, egg number, sperm velocity, and sperm longevity. Do sperm
traits vary with egg traits as selection for increased egg size proceeds? How correlated
are sperm traits and egg traits within individuals, within the population?
The observed variation in size of males suggests additional work on sex change in
this animal. It is reported that all individuals change sex, but many males remaining in
the population at 6 weeks were considerably larger than females. Allsop and West
(2003) showed that the relative timing of sex change was invariant across broad
taxonomic groups and largely predicted by maximum body size, that is, that a sex change
should occur at 72% of maximum body size. Data collected in this study do not provide
a way to test this prediction but population level studies would be a useful approach to
this question. This same data set could also inform us concerning the genetic variation
for maximum body size and the trade-off between growth and fecundity at the population
Reznick et al. (2000) pointed out that studies evaluating life-history traits (and the
trade-offs between them) necessarily focus on specific characters. The choice of what
components of fitness are included in a study can make the difference between detecting
a fitness trade-off or not, since the influence of correlations between unmeasured traits is
not known. As efforts to examine the genetic basis of trade-offs broaden to include
marine invertebrate life-history evolution, methods that are the most integrative across
the life cycle (including various developmental stages/ages/sizes) and across the breadth
of traits that define a life history may be the most effective. Large sample sizes are often
required to detect genetic correlations using breeding designs and this is reasonable in
insect systems. Frequently in marine invertebrate systems this is a logistical
impossibility because culture of the larvae is time-, space-, and labor-intensive. It is
difficult to detect complex or unexpected patterns in studies that examine correlations
across a single generation when sample sizes become limiting. These patterns may be
more apparent against a dynamic background where observations are made across
multiple life-history stages and generations.
Gen 6 Gen 7 Gen 8 Gen 9 Gen 10 Gen 11
Figure 3-1. Cumulative response to selection for increased egg diameter in H. elegans
was positive through Generation 11. No error bars were calculated because
this is the difference between the mean of 3 selected lines and the mean of 3
control lines at each generation beginning in the base population (Generation
6) and ending at the end of selection (Generation 11).
Table 3-1. Egg diameter ([tm) in replicate selected vs. control lines ofHydroides elegans
after 4 generations of selection; Fully nested ANOVA
Source df EMS (X 10-4) F P
Treatment 1 147.0604 95.7328 0.0006
Line (Treat) 4 01.5686 6.1680 0.0001
Cross (Line (Treat)) 428 0.2545 18.1511 0.0001
Error 3916 0.0140 -- --
Table 3-2. Total egg energy in replicate selected vs. control lines ofHydroides elegans
(calculated as |tg C egg-'); Fully nested ANOVA
Source df EMS (X 10-4) F P
Treatment 1 2.535578 9.8169 0.0350
Line (Treat) 4 0.258603 2.3369 0.0611
Cross (Line (Treat)) 90 0.111605 1.6279 0.0032
Error 174 0.0686 -- --
Table 3-3. Absolute fecundity in replicate selected vs. control lines ofHydroides elegans
(eggs female-'); Fully nested ANOVA
Source df EMS (X 106) F P
Treatment 1 443.82396 10.5232 0.0304
Line (Treat) 4 42.662385 1.9105 0.1078
Error 403 22.330157 ---
Table 3-4. Relative fecundity based on dry weight of the female in replicate selected vs.
control lines ofHydroides elegans (eggs mg-1); Fully nested ANOVA
Source df EMS (X 106) F P
Treatment 1 795.873456 6.0931 0.0679
Line (Treat) 4 132.548182 2.5540 0.0386
Error 403 51.897613 -- -
Table 3-5. Adult female dry weight (mg) in replicate selected vs. control lines of
Hydroides elegans; Fully nested ANOVA
Source df EMS F P
Treatment 1 0.025218 0.2290 0.6567
Line (Treat) 4 0.112206 2.8075 0.0253
Error 432 0.0399669 -- -
Table 3-6. Larval volume ([tL) at 18 to 20 h post-fertilization in replicate selected vs.
control lines ofHydroides elegans; Fully nested ANOVA
Source df EMS (X 10-4) F P
Treatment 1 0.650836 1.1703 0.3392
Line (Treat) 4 0.558838 1.5227 0.2048
Cross (Line (Treat)) 69 0.366555 0.9246 0.6495
Error 648 0.3964 -- -
Table 3-7. Volume ([tL) of competent (5 d) larvae in replicate selected vs. control lines
ofHydroides elegans; Fully nested ANOVA
Source df EMS (X 10-6) F P
Treatment 1 2.44511 4.9009 0.0894
Line (Treat) 4 0.50505 1.8355 0.1324
Cross (Line (Treat)) 66 0.276535 4.9538 0.0001
Error 641 0.0006 .
Table 3-8. Juvenile tube length (mm) at 21 d in replicate selected vs. control lines of
Hydroides elegans; Fully nested ANOVA
Source df EMS F P
Treatment 1 47.0398 3.7050 0.1249
Line (Treat) 4 12.6558 0.6682 0.6162
Cross (Line (Treat)) 69 19.16195 4.7420 0.0001
Error 558 4.0409
Table 3-9. Summary of direct and correlated responses to selection for increased egg
diameter in replicate selected v controls lines ofHydroides elegans. All
significant differences represent positive correlations
Trait Selected Mean (+ SE) Control Mean (+ SE)
Egg d (im) 49.23 (0.35) 45.46 (0.083) *
Total Energy 0.37 (0.17) 0.29 (0.18) *
18 to 20 h larval
5 d larval Vol
21 d Juvenile tube
1.12 X10-4 (0.064 X 10-4)
1.29 X 10-3 (0.041 X 10-3)
Adult female dry 0.80 (0.019)
Fecundity (eggs) 10071 (481)
Relative 13451 (834)
* P < 0.05.+P < 0.10. NSP>0.10
1.14 X10-4 (0.047 X 10-4)
1. 16 X 10-3 (0.036 X 10-3)
Table 3-10. Effect of selection (treatment) on the relative numbers of males and females
in Generations 7 to 11; Model I ANOVA
Source df EMS F P
Treatment 1 1.3958 6.28 0.0122
Gen 4 0.3852 1.73 0.1395
Gen*treat 4 0.1717 0.77 0.5427
Error 3461 0.2221 -- -
Table 3-11. Mean percent females present at spawning (6 weeks post-fertilization) in
replicate selected vs. control lines ofHydroides elegans in Generations 7 to 11
Mean Percent Females in Mean Percent Females in
Selected Lines (SE) Control Lines (SE)
Generation 7 64.70 (9.036) 61.86 (2.281)
Generation 8 64.88 (2.299) 59.21 (2.291)
Generation 9 72.11 (5.564) 61.72 (0.877
Generation 10 70.45 (9.022) 62.61 (6.820
Generation 11 74.30 (8.661) 64.21 (3.071
Grand Mean 69.29 (1.935) 61.92 (0.809
The goal of this study was to investigate the evolvability of the trait of egg size, to
examine the direct and correlated responses to selection on increased egg size in a
specified set of life-history characters, and to explore how these correlations might
change as egg size was increased by artificial selection in the polychaete worm
Hydroides elegans. Life-history traits examined included egg diameter, fecundity,
relative fecundity, total egg energy (at 6 weeks), larval size at 18 to 20 h, larval size at
competence (5 d), tube length at 21 d, and adult dry weight (at 6 weeks). I have used
various analytical techniques to show that there is significant additive genetic variation
for egg size in the Pearl Harbor population ofH. elegans. The response of 2.5 op to
direct selection on egg size indicates that there is substantial potential for egg size to
increase in response to selective pressures (Chapter 2). I predicted that there would be
negative correlations between at least some fitness-related traits, because we expect
alleles underlying positive genetic correlations to be rapidly fixed in natural populations.
Yet, I found that large-egg selected females produced more eggs, with higher energy
content, that grew into larger larvae at competence. I have proposed that the constraint
on increased egg diameter (at 6 weeks) in the Pearl Harbor population of Hydroides
elegans appears to be selective pressure to maintain the optimum male contribution to
lifetime reproduction by delaying the shift to the female sex (Chapter 3).
Phenotypic correlations between the trait of egg diameter and the suite of
fitness-related traits outlined above were estimated using bivariate analysis on replicate
line means (n = 3 lines treatment-) in JMP 5.1 (SAS Institute, Inc.). The results
comparing control with selected line means can be seen as Pearson product moment
correlations (r) in Table 4-1.
Among lines the mean values for controls showed no phenotypic correlation
between egg diameter and any of the traits of interest, with the exception of larval size at
competence (5 d) where a significant positive correlation (0.98) was found. This
relationship persisted in selected line means (0.99). However, despite the analysis at the
level of treatment showing positive correlations between egg diameter and fecundity,
examination of individual selected lines showed a significant negative correlation
between egg diameter and both measures of fecundity (-0.95 and -0.99 respectively). The
selected lines also showed a significant positive correlation between egg diameter and
female dry weight (0.99). A pattern is suggested in the differences between control and
selected lines with respect to these phenotypic correlations; the replicate selected line that
achieved the highest mean egg diameter also had the highest female dry weight and the
lowest fecundity. This pattern (the largest mothers producing the largest eggs in the
fewest numbers) is reminiscent of the predicted trade-off between growth, egg size, and
fecundity that I did not detect between treatments (Chapter 3). This relationship emerges
at the extreme end of the continuum of reproductive investment (large-egg selected lines)
and is not detectable among the control lines. While this observation cannot be
considered quantitative (n = 3 is the minimum sample size for a correlation), it suggests
that further study into the influence of trade-offs in defining reproductive investment
might be most fruitful at the extreme ends of the investment continuum in this species.
Within polychaetes, fecundity within species can vary with nutritional state of the
mother, population density, and adult age and size (Eckelbarger 1986). In this
experiment all animals were fed ad libidum at all stages of development, while density
and adult age were controlled. Varying any of these parameters or inducing
environmental stress at different developmental stages might also reveal evidence of
trade-offs influencing reproductive effort. Qiu and Qian (1998) looked at effects of
varying salinity and temperature on juvenile traits in H. elegans, and found that fecundity
was affected by the stress of low salinity. Czesak and Fox (2003) found while working
with the seed beetle Stator limbatus that the magnitude of the tradeoff between egg size
and fecundity varied between environments.
Another possible cost associated with increased reproductive effort is higher
mortality, as was found in the polychaete Dinophiilus gyrociliatus by Prevedelli and
coworkers (Prevedelli and Zunarelli Vandini 1999, Prevedelli and Simonini 2000). They
found that higher food led to increased fecundity but was accompanied by greater and
earlier mortality. Tatar et al. (1993) nicely demonstrated the tradeoff between increased
egg production and increased age-specific mortality in the beetle Callosobruchus
Many avenues of inquiry can be pursued based on this study. A large body of work
has been compiled investigating the adaptive significance of developmental mode in
marine invertebrates using comparative, empirical, and manipulative techniques. A great
deal of effort has also been invested in quantitative modeling approaches that have been
used to try and identify selection pressures and processes that influence developmental
mode evolution. The suite of characteristics that defines a developmental type has
co-evolved in concert and compromise with a wider complex of life-history traits under
the influence of natural selection. The most rewarding efforts to pursue these dynamic
and sometimes subtle relationships will be methods that are the most integrative across
the stages that define a life cycle, and across the breadth of traits that define a life history.
Understanding the nature of the variation in fitness-related traits and how it is partitioned
in a particular population has the potential to reveal the insights we seek concerning the
evolution of the diverse life histories among marine invertebrates.
Table 4-1. Phenotypic correlations between the trait of egg size and a suite of life-history
traits in replicate selected and control lines in the polychaete Hydroides
elegans, based on line means (n = 3 lines treatment-1). Pearson product
moment correlations (r) with absolute value > 0.95 indicated with *
Correlation coefficient (r) among Correlation coefficient (r) among
Control Line Means (n = 3) Selected Line Means (n = 3)
Egg d Fecundity 0.09 Egg d Fecundity -0.95*
Relative Relative -0.99*
Egg d 0.24 Egg d
Egg d Female Dry -0.05 Egg Female Dry 0.99*
Egg d Total energy -0.48 Egg d Total energy -
Egg d 1 -0.48 Egg d 1 -0.28
Larval size Larval size
gg at 18 to 20 h ggat 18 to 20 h
Sd Larval size .* Larval size 0.
Eggd at 5 d 0.98* Egg d at5d 0.99
at 5d at 5d
Sd Tube length 6 E Tube length -
d at21d -86 d at 21 d -
LIST OF REFERENCES
Allan, J.D. 1984. Life history variation in a freshwater copepod: evidence from
population crosses. Evolution 38:280-291.
Allsop, D.J., and S.A. West. 2003. Changing sex at the same relative body size. Nature
Arcos, F.G., I.S. Racotta, and A.M. Ibarra. 2004. Genetic parameter estimates for
reproductive traits and egg composition in Pacific white shrimp Penaeus
(Litopenaeus) vannamei. Aquaculture 236:151-165.
Azevedo, R.B.R., V. French, and L. Partridge. 1996. Thermal evolution of egg size in
Drosophila melanogaster. Evolution 50:2338-2345.
Azevedo, R.B.R., V. French, and L. Partridge. 1997. Life-history consequences of egg
size in Drosophila melanogaster. American Naturalist. 150:250-282.
Bakker, K. 1969. Selection for rate of growth and its influence on competitive ability of
larvae of Drosophila melanogaster. Netherlands Journal of Zoology. 19:541-595.
Barnes, H., and M. Barnes. 1965. Egg size, nauplius size, and their variation with local,
geographical and specific factors in some common cirripedes. Journal of Animal
Bell, A.E., C.H. Moore, and D.C. Warren. 1955. The evaluation of new methods for the
improvement of quantitative characteristics. Cold Spring Harbor Symposia on
Quantitative Biology. 20:197-212.
Berglund, A. 1990. Sequential hermaphroditism and the size-advantage hypothesis: an
experimental test. Animal Behavior 39:426-433.
Bemardo, J. 1996. The particular maternal effect of propagule size, especially egg size:
patterns, models, quality of evidence and interpretations American Zoologist
Blanckenhorn, W.U., and A. Heyland. 2004. The quantitative genetics of two life history
trade-offs in the yellow dung fly in abundant and limited food environments
Evolutionary Ecology 18:385-402.
Bouchet, P. 1989. A review of poecilogony in gastropods. Journal of Molluscan Studies
Bradshaw, W.E., and C.M. Holzapfel. 1996. Genetic constraints to life-history evolution
in the pitcher-plant mosquito, Wyeomyia smithii. Evolution 50:1176-1181.
Brakefield, P.M. 2003. Artificial selection and the development of ecologically relevant
phenotypes. Ecology 84:1661-1671.
Bridges, T.S., and S. Heppell. 1996. Fitness consequences of maternal effects in
Streblospio benedicti (Annelida: Polychaeta). American Zoologist 36:132-146.
Carpizio-Ituarte, E., and M.G. Hadfield. 1998. Stimulation of metamorphosis in the
polychaete Hydroides elegans Haswell (Serpulidae). Biological Bulletin 194:14-24.
Caswell, H. 1981. The evolution of "mixed" life histories in marine invertebrates and
elsewhere. American Naturalist 117:529-536.
Charnov, E.L. 1979a. Natural selection and sex change in pandalid shrimp: a test of a life
history theory. The American Naturalist 113:715-734.
Charnov, E.L. 1979b. The genetical evolution of patterns of sexuality: Darwinian fitness.
The American Naturalist 113:465-480.
Charnov, E.L. 1982. The theory of sex allocation. Princeton University Press. Princeton,
NJ. 355 p.
Cheverud, J.M. 1984. Quantitative genetics and developmental constraints on evolution
by selection. Journal of Theoretical Biology 110:155-171.
Chia, F.-S. 1974. Classification and adaptive significance of developmental patterns in
marine invertebrates. Thalassia Jugoslavica 10:121-130.
Christiansen, F.B., and T.M. Fenchel. 1979. Evolution of marine invertebrate
reproductive patterns. Theoretical Population Biology 16:267-282.
Clarke, A. 1993. Egg size and egg composition in polar shrimps (Caridea; Decapoda)
Journal of Experimental Marine Biology and Ecology 168:189-203.
Collin, R. 1995. Sex, size, and position: a test of models predicting size at sex change in
the protandrous gastropod Crepidulafornicata. The American Naturalist 146:815-
Collin, R. 2001. Mode of development, phylogeography, and population structure of
North Atlantic Crepidula (Gastropoda: Calyptraeidae) species. Molecular Ecology
Cunningham, C.W. 1999. Some limitation of ancestral character-state reconstruction
when testing evolutionary hypotheses. Systematic Biology 48:665-674.
Czesak, M.E., and C.W. Fox. 2003. Evolutionary ecology of egg size and number in a
seed beetle: genetic trade-off differs between environments. Evolution 57:1121-
Duda, T.F., and S.R. Palumbi. 1997. Developmental shifts and species selection in
gastropods. Proceedings of the National Academy of Science USA 96:10272-
Ebert, D. 1993. The trade-off between offspring size and number in Daphnia magna: the
influence of genetic, environmental and maternal effects. Archiv fur Hydrobiologie
Eckelbarger, K.J. 1986. Vitellogenic mechanisms and the allocation of energy to
offspring in polychaetes. Bulletin of Marine Science 39:426-443.
Edmands, S., and J.S. Harrison. 2003. Molecular and quantitative trait variation within
and among populations of the intertidal copepod Tigriopus californicus. Evolution
Emlet, R.B. 1995. Developmental mode and species geographic range in regular sea
urchins (Echinodermata: Echinoidea). Evolution 49:476-489.
Emlet, R.B., and 0. Hoegh-Guldberg. 1997. Effects of egg size on postlarval
performance: experimental evidence from a sea urchin. Evolution 51:141-152.
Emlet, R.B., L.R. McEdward, and R.R. Strathmann. 1987. Echinoderm larval ecology
viewed from the egg. pp. 55-136. In: Jangoux, M. and J.M. Lawrence (eds.) 1987.
Echinoderm studies. Balkema. Rotterdam.
Englert, D.C., and A.E. Bell. 1969. Components of growth in genetically diverse
populations of Tribolium castaneum. Canadian Journal of Genetics and Cytology
Englert, D.C., and A.E. Bell. 1970. Selection for time of pupation in Tribolium
castaneum. Genetics 64:541-552.
Ernande, B., P. Boudry, J. Clobert, and J. Haure. 2004. Plasticity in resource allocation
based life history traits in the Pacific oyster, Crassostrea gigas. I. Spatial variation
in food abundance. Journal of Evolutionary Biology 17:342-356.
Ernande, B., J. Clobert, H. McCombie, and P. Boudry. 2003. Genetic polymorphism and
trade-offs in the early life-history strategy of the Pacific oyster, Crassostrea gigas
(Thunberg, 1795): a quantitative genetic study. Journal of Evolutionary Biology
Falconer, D.S., and T.F.C. Mackay. 1996. Introduction to quantitative genetics. Prentice
Hall. London. 464 p.
Fischer, K., B.J. Zwaan, and P.M. Brakefield. 2004. Genetic and environmental sources
of egg size variation in the butterfly Bicyclus anynana. Heredity. 92:163-169.
Fox, C.W. 1993. Maternal and genetic influences on egg size and larval performance in a
seed beetle (Callosobruchus maculatus): multigenerational transmission of a
maternal effect? Heredity. 73:509-517.
Fox, C.W., and M.W. Czesak. 2000. Evolutionary ecology of progeny size in arthropods.
Annual Review of Entomology 45:341-369.
George, S.B. 1999. Egg quality, larval growth and phenotypic plasticity in a forcipulate
seastar. Journal of Experimental Marine Biology and Ecology 237:203-224.
Ghiselin, M.T. 1969. The evolution of hermaphroditism among animals. Quarterly
Review of Biology 44:189-208.
Gibson, G.D. 1993. Developmental variability and the induction of metamorphosis in
Haminaea callidegenita (Mollusca: Opisthobranchia). Zoology. Edmonton,
Alberta, University of Alberta: 134 p.
Gimenez, L. 2002. Effects of prehatching salinity and initial larval biomass on survival
and duration of development in the zoea 1 of the estuarine crab, Chasmagnathus
granulata. Journal of Experimental Marine Biology and Ecology 270:93-110.
Gimenez, L., and K. Anger. 2003. Larval performance in an estuarine crab,
Chasmagnathus granulata, is a consequence of both larval and embryonic
experience. Marine Ecology Progress Series 249:251-264.
Gimenez, L., K. Anger, and G. Torres. 2004. Linking life history traits in successive
phases of a complex life cycle: effects of larval biomass on early juvenile
development in an estuarine crab, Chasmagnathus granulata. Oikos 104:570-580.
Gosselin, L.A., and P.-Y. Qian. 1999. Analysing energy content: a new micro-assay and
an assessment of the applicability of acid dichromate assays. Hydrobiologia
Grant, A. 1983. On the evolution of brood protection in marine benthic invertebrates.
American Naturalist 122:549-555.
Guillard, R.R.L. 1975. Culture of phytoplankton for feeding marine invertebrates. pp. 29-
60. In: Smith, W.L. and M.H. Chaney (eds.). Culture of Marine Invertebrate
Animals. 1975. Plenum Publishing, New York.
Gustafsson, L. 1986. Lifetime reproductive success and heritability: empirical support for
Fisher's fundamental theorem. American Naturalist 128:761-764.
Hadfield, M.G., and S.E. Miller. 1987. On developmental patterns of opisthobranchs.
American Malacological Bulletin 5:197-214.
Hadfield, M.G., and M.F. Strathmann. 1996. Variability, flexibility, and plasticity in life
histories of marine invertebrates. Oceanologica Acta 19:323-334.
Hadfield, M.G., C.C. Unabia, C.M. Smith, and T.M. Michael. 1994. Settlement
preferences of the ubiquitous fouler Hydroides elegans. pp. 65-74. In: Thompson,
M, R. Nagabhushanam, R. Sarojini, and M. Fingerman (eds.). Recent
Developments in Biofouling Control. 1994. Oxford and IBH Publishing Co. Pvt.
Ltd. New Delhi.
Harshman, L.G., and A.A. Hoffmann. 2000. Laboratory selection experiments using
Drosophila: what do they really tell us? Trends in Ecology and Evolution 15:32-36.
Hart, M.W. 1995. What are the costs of small egg size for a marine inverterate with
feeding planktonic larvae? American Naturalist 146:415-426.
Hart, M.W. 1996. Evolutionary loss of larval feeding: development, form and function in
a facultatively feeding larva, Brisaster latifrons. Evolution 50:174-187.
Harvey, G.T. 1983. Environmental and genetic effects on mean egg weight in Spruce
Budworm (Lepidoptera: Tortricidae) Canadian Entomologist 115:1109-1117.
Havenhand, J.N. 1993. Egg to juvenile period, generation time, and the evolution of
larval type in marine invertebrates. Marine Ecology Progress Series 97:247-260.
Havenhand, J.N. 1995. Evolutionary ecology of larval types. pp. 79-122. In: McEdward,
L.R. (ed). 1995. Ecology of marine invertebrate larvae. CRC Press. Boca Raton.
Hedgecock, D. 1982. Is gene flow from pelagic larval dispersal important in the
adaptation and evolution of marine invertebrates? Bulletin of Marine Science
Hendry, A.P., T. Day, and A.B. Cooper. 2001. Optimal size and number of propagules:
allowance for discrete stages and effects of maternal size on reproductive output
and offspring fitness. American Naturalist 157:387-407.
Hilbish, T.J., E.P. Winn, and P.D. Rawson. 1993. Genetic variation and covariation
during larval and juvenile growth in Mercenaria mercenaria. Marine Biology
Hill, W.G. 1971. Design and efficiency of selection experiments for estimating genetic
parameters. Biometrics 27:293-311.
Hill, W.G. 1972. Estimation of realized heritabilities from selection experiments. II.
Selection in one direction. Biometrics 28:767-780.
Hoagland, K.E., and R. Robertson. 1988. An assessment of poecilogony in marine
invertebrates: phenomenon or fantasy? Biological Bulletin 174:109-125.
Hohls T. 1998. Reliability of confidence interval estimators under various nested design
parental sample sizes. Biometrical Journal 40:85-98.
Hoskin, M.G. 1997. Effects of contrasting modes of larval development on the genetic
structure of 3 species of prosobranch gastropods. Marine Biology 127:647-656.
Houle, D. 1992. Comparing evolvability and variability of quantitative traits. Genetics
Jablonski, D. 1986. Larval ecology and macroevolution in marine invertebrates. Bulletin
of Marine Science 39:565-587.
Jablonski, D. 1987. Heritability at the species level: analysis of geographic ranges of
Cretaceous mollusks. Science 238:360-363.
Jaeckle, W.B. 1995. Variation in the size, energy content and biochemical composition of
invertebrate eggs: correlates to the mode of larval development. pp. 49-77. In:
McEdward, L.R. (ed). 1995. Ecology of marine invertebrate larvae. CRC Press.
Boca Raton, FL.
Jagersten, G. 1972. Evolution of the metazoan life cycle: a comprehensive theory.
Academic Press. London. 282 p.
Jarrett, J.N., and J.A. Pechenik. 1997. Temporal variation in cyprid quality and juvenile
growth capacity for an intertidal barnacle. Ecology 78:1262-1265.
Jones, H.L., C.D. Todd, and W.J. Lambert. 1996. Intraspecific variation in embryonic
and larval traits of the dorid nudibranch mollusc Adalaria proxima (Alder and
Hancock) around the northern coasts of the British Isles. Journal of Experimental
Marine Biology and Ecology 202:29-47.
Knapp S.J. 1986. Confidence intervals for heritability for two-factor mating design single
environment linear models. Theoretical Applied Genetics 72:587-591.
Kohn, A.J., and F.E. Perron. 1994. Life history and biogeography: patterns in Conus.
Oxford University Press. New York. 122 p.
Koots, K.R., and J.P. Gibson. 1996. Realized sampling variances of estimates of genetic
parameters and the difference between genetic and phenotypic correlations.
Kupriyanova, E.K., E. Nishi, H.A. ten Hove, and A.V. Rzhavsky. 2001. Life-history
patterns in serpulimorph polychaetes. Annual Review of Oceanography and Marine
Laughlin, R.B., and W. French. 1989. Interactions between temperature and salinity
during brooding on subsequent zoeal development of the mud crab
Rhithropanopeus harrisii Marine Biology 102:377-386.
Leigh, E.G., E.L. Charnov, and R.R. Warner. 1976. Sex ratio, sex change, and natural
selection. Proceedings of the National Academy of Science, USA 73:3656-3660.
Leonardos, N., and I.A.N. Lucas. 2000. The nutritional value of algae grown under
different culture conditions for Mytilus edulis L. larvae. Aquaculture 182:301-315.
Lessios, H.A. 1987. Temporal and spatial variation in egg size of 13 Panamanian
echinoids. Journal of Experimental Marine Biology and Ecology 114:217-239.
Levin, L.A., H. Caswell, K.D. DePatra, and E.L. Creed. 1987. Demographic
consequences of larval development mode: planktotrophy vs. lecithotrophy in
Streblospio benedicti. Ecology 68:1877-1886.
Levin, L.A., and E.L. Creed. 1986. Effect of temperature and food availability on
reproductive responses of Streblospio bendicti (Polychaeta: Spionidae) with
planktotrophic or lecithotrophic development. Marine Biology 92:103-113.
Levin, L.A., and D.V. Huggett. 1990. Implications of alternative developmental modes
for the seasonal dynamics and demography of an estuarine polychaete. Ecology
Levin, L.A., J. Zhu, and E.L. Creed. 1991. The genetic basis of life-history characters in a
polychaete exhibiting planktotrophy and lecithotrophy. Evolution 45:380-397.
Levitan, D.L. 1993. The importance of sperm limitation to the evolution of egg size in
marine invertebrates. The American Naturalist 141:517-536.
Levitan, D.L. 1996. Predicting optimal and unique egg sizes in free-spawning marine
invertebrates. The American Naturalist 148:174-188.
Levitan, D.L. 1998. Sperm limitation, sperm competition and sexual selection in external
fertilizers. pp. 173-215. In: Birkhead, T.R., and A.P. Moller (eds.). Sperm
competition and sexual selection. 1998. Academic Press. New York.
Levitan, D.R. 2000. Optimal egg size in marine invertebrates: theory and phylogenetic
analysis of the critical relationship between egg size and development time in
echinoids. American Naturalist 156:175-192.
Lips, K.R. 2001. Reproductive trade-offs and bet-hedging in Hyla calypsa, a neotropical
treefrog. Oecologia 128:509-518.
Luttikhuizen, P. C., P.J.C. Honkoop, J. Drent, and J. van der Meer. 2004. A general
solution for optimal egg size during external fertilization, extended scope for
intermediate optimal egg size and the introduction of don Ottavio 'tango'. Journal of
Theoretical Biology 231:333-343.
Lynch, M. 1984. The limits to life history evolution in Daphnia. Evolution 38:465-482.
Lynch, M., and B. Walsh. 1998. Genetics and Analysis of Quantitative Traits. Sinauer
Associates, Inc. Sunderland, MA. 979 p.
Maldonado, M., and C.M. Young. 1999. Effects of the duration of larval life on
postlarval stages of the demosponge Sigmadocia caerulea. Journal of Experimental
Marine Biology and Ecology 232:9-21.
Markow T.A., G.A. Clarke. 1997. Meta-analysis of the heritability of developmental
stability: a giant step backward. Journal of Evolutionary Biology 10:31-37.
Marshall, D.J., C.N. Cook, and R.B. Emlet. 2006. Offspring size effects mediate
competitive interactions in a colonial marine invertebrate. Ecology 87:214-225.
Marshall, D.J., C.A. Styan, and M.J. Keough. 2002. Sperm environment affects offspring
quality in broadcast spawning marine invertebrates. Ecology Letters 5:173-176.
McEdward, L.R. 1997. Reproductive strategies of marine benthic invertebrates revisited:
facultative feeding by planktotrophic larvae. American Naturalist 150:48-72.
McEdward, L.R., and S.F. Carson. 1987. Variation in egg organic content and its
relationship with egg size in the starfish Solaster stimpsoni. Marine Ecology
Progress Series 37:159-168.
McEdward, L.R., and L.K. Coulter. 1987. Egg volume and energetic content are not
correlated among sibling offspring of starfish: implications for life-history theory.
McEdward, L.R., and B.G. Miner. 2001. Larval and life cycle patterns in echinoderms.
Canadian Journal of Zoology 79:1125-1170.
McMillian, W.O., R.A. Raff, and S.R. Palumbi. 1992. Population genetic consequences
of developmental evolution in sea urchins (genus Heliocidaris). Evolution 46:1299-
Miner, B.G., L.A. McEdward, and L.R. McEdward. 2005. The relationship between egg
size and the duration of the facultative period in marine invertebrate larvae. Journal
of Experimental Marine Biology and Ecology. 321:135-144.
Moran, N.A. 1994. Adaptation and constraint in the complex life cycles of animals
Annual Review of Ecology and Systematics 25:573-600.
Moran, A.L., and R.B. Emlet. 2001. Offspring size and performance in variable
environments: field studies on a marine snail. Ecology 82:1597-1612.
Moran, N.A. 1994. Adaptation and constraint in the complex life cycles of animals
Annual Review of Ecology and Systematics 25:573-600.
Mousseau, T.A., and C.W. Fox. 1998. Maternal effects as adaptations. Oxford University
Press. New York. 375 p.
Mousseau, T.A., D.A. Roff. 1987. Natural selection and the heritability of fitness
components. Heredity 59:181-197.
Nunney, L. 1996. The response to selection for fast larval development in Drosophila
melanogaster and its effect on adult weight: an example of a fitness trade-off.
Palmer, J.O., and H. Dingle. 1986. Direct and correlated responses to selection among
life-history traits in milkweed bugs (Oncopeltusfasciatus). Evolution 40:767-777.
Parsons, P. 1964. Egg lengths in Drosophila melanogaster and correlated responses to
selection. Genetica 35:175-181.
Parsons, T.R., Y. Maita, and C.M. Lalli. 1984. A manual of chemical and biological
methods for seawater analysis. Pergamon Press. Oxford. 173 p.
Partridge, L., and K. Fowler. 1992. Direct and correlated responses to selection on age at
reproduction in Drosophila melanogaster. Evolution 46:76-91.
Pechenik, J.A. 1999. On the advantages and disadvantages of larval stages in benthic
marine invertebrate life cycles. Marine Ecology-Progress Series 177:269-297.
Pechenik, J.A., and T.R. Cerulli. 1991. Influence of delayed metamorphosis on survival,
growth, and reproduction of the marine polychaete Capitella-sp.-I. Journal of
Experimental Marine Biology and Ecology 151:17-27.
Pechenik, J.A., T.J. Hilbish, L.S. Eyster, and D. Marshall. 1996. Relationship between
larval and juvenile growth rates in two marine gastropods, Crepidulaplana and C.
fornicata. Marine Biology 125:119-127.
Pechenik, J.A., and M.E. Rice. 2001. Influence of delayed metamorphosis on
postsettlement survival and growth in the sipunculan Apionsoma misakianum.
Invertebrate Biology 120:50-57.
Pechenik, J.A., D. Rittschof, and A.R. Schmidt. 1993. Influence of delayed
metamorphosis on survival and growth of juvenile barnacles Balanus amphitrite.
Marine Biology 115:287-294.
Pechenik, J.A., D.E. Wendt, and J.N. Jarrett. 1998. Metamorphosis is not a new
beginning. Bioscience 48:901-910.
Pernet, B., and W.B. Jaeckle. 2004. Size and organic content of eggs of marine annelids,
and the underestimation of egg energy content by dichromate oxidation. Biological
Perron, F.E., and R.H. Carrier. 1981. Egg size distributions among closely related marine
invertebrate species: are they bimodel or unimodel? American Naturalist 118:749-
Phillips, N.E. 2002. Effects of nutrition-mediated larval condition on juvenile
performance in a marine mussel. Ecology 83:2562-2574.
Pinheiro, J., and D. Bates. 2000. Mixed-effects models in S and S-plus. Springer-Verlag.
New York. 528 p.
Podolsky, R.D., and R.R. Strathmann. 1996. Evolution of egg size in free-spawners:
consequences of the fertilization-fecundity trade-off. American Naturalist 148:160-
Prevedelli, D., and R. Simonini. 2000. Effecs of salinity and two food regimes on
survival, fecundity and sex ratio in two groups ofDinophilus gyrociliatus
(Polychaeta: Dinophilidae). Marine Biology 137:23-29.
Prevedelli, D., and R. Zunarelli Vandini. 1999. Survival, fecundity and sex ratio of
Dinophilus gyrociliatus (Polychaeta: Dinophilidae) under different dietary
conditions. Marine Biology 133:231-236.
Qian, P.-Y., and J.A. Pechenik. 1998. Effects of larval starvation and delayed
metamorphosis on juvenile survival and growth of the tube-dwelling polychaete
Hydroides elegans. Journal of Experimental Marine Biology and Ecology 227:169-
Qian, P.-Y., and F.-S. Chia. 1991. Fecundity and egg size are mediated by food quality in
the polychaete worm Capitella sp. Journal of Experimental Marine Biology and
Qiu, J.-W., and P.-Y. Qian. 1997. Combined effects of salinity, temperature and food on
early development of the polychaete Hydroides elegans. Marine Ecology Progress
Qiu, J-W., and P.-Y. Qian. 1998. Combined effects of salinity and temperature on
juvenile survival, growth and maturation in the polychaete Hydroides elegans.
Marine Ecology Progress Series 168:127-134.
Raff, R. 1987. Constraint, flexibility, and phylogenetic history in the evolution of direct
development in sea urchins. Developmental Biology 119:6-19.
Ramirez Llodra, E. 2002. Fecundity and life-history strategies in marine invertebrates.
pp. 87-170. In: Southward, A.J., P.A. Tyler, C.M. Young, and L.A. Fuiman (eds.).
2002. Advances in Marine Biology. Academic Press. Amsterdam.
Ranzoli, F. 1962. II falso gonocorismo nel ciclo biologico del polichete Hydroides
norvegica (Gunn.). Archivio Zoologico 47:1-19.
Reznick, D. 1982. The impact of predation on life history evolution in Trinidadian
guppies: genetic basis of observed life history traits. Evolution 36:1236-1250.
Reznick, D. 1985. Costs of reproduction: an evaluation of the empirical evidence. Oikos
Reznick, D., L. Nunney, and A. Tessier. 2000. Big houses, big cars, superfleas and the
costs of reproduction. Trends in Ecology and Evolution 15:421-425.
Roff, D.A. 1990. Understanding the evolution of insect life-cycles: the role of genetic
analysis. pp. 5-27. In: Gilbert, F. (ed.). Insect life-cycles: genetics, evolution and
co-ordination. 1990. Springer Verlag. London.
Roff, D.A. 1992. The evolution of life histories: theory and analysis. Chapman and Hall.
New York. 535 p.
Roff, D.A., and T.A. Mousseau. 1987. Quantitative genetics and fitness: lessons from
Drosophila. Heredity 58:103-118.
Rose, M.R., J.L. Graves, and E.W. Hutchinson. 1990. The use of selection to probe
patterns of pleiotropy in fitness characters. pp. 29-42. In: Gilbert, F. (ed.). Insect
life-cycles: genetics, evolution and co-ordination. 1990. Springer Verlag. London.
Roughgarden, J. 1989. The evolution of marine life cycles. pp. 270-330. In: Feldman,
M.W. (ed.). 1989. Mathematical evolutionary theory. Princeton University Press.
Rouse, G., and K. Fitzhugh. 1994. Broadcasting fables: is external fertilization really
primitive? Sex, size, and larvae in sabellid polychaetes. Zoologica Scripta 23:271-
Rouse, G.W. 2000. Bias? What bias? The evolution of downstream larval-feeding in
animals. Zoologica Scripta 29:213-236.
Scheltema, R.S. 1989. Planktonic and non-planktonic development among prosobranch
gastropods and its relationships to the geographic ranges of species. pp. 183-188.
In: Ryland, J.S., and P.A. Tyler (eds.). Reproduction, genetics, and distributions of
marine organisms. Olsen and Olsen. Fredensborg, Denmark.
Schwarzkopf, L., M.W. Blows, and M.J. Caley. 1999. Life-history consequences of
divergent selection on egg size in Drosophila melanogaster. American Naturalist.
Seslija, D., and N. Tucic. 2003. Selection for developmental time in bean weevil
(.-L iu//1t/1/,L li d'\ obtectus): correlated responses for other life history traits and
genetic architecture of line differentiation. Entomologia Experimentalis et
Sewell, M.A. 1994. Small size, brooding, and protandry in the apodid sea cucumber
Leptosynapta clarki. Biological Bulletin 187:112-123.
Shaw, R.G. 1987. Maximum-likelihood approaches applied to quantitative genetics of
natural populations. Evolution 41:812-826.
Sinervo, B. 1990. The evolution of maternal investment in lizards: an experimental and
comparative analysis of egg size and its effects on offspring performance.
Sinervo, B., and L.R. McEdward. 1988. Developmental consequences of an evolutionary
change in egg size: an experimental test. Evolution 42:885-899.
Smith, C.C., and S.D. Fretwell. 1974. The optimal balance between egg size and number
of offspring. American Naturalist 108:499-506.
Stearns, S.C. 1983. The genetic basis of differences in life-history traits among six
populations of mosquitofish (Gambusia affinis) that shared ancestors in 1905.
Stearns, S.C. 1992. The evolution of life histories. Oxford University Press. Oxford.
Strathmann, R.R. 1978. The evolution and loss of feeding larval stages of marine
invertebrates. Evolution 32:894-906.
Strathmann, R.R. 1985. Feeding and non-feeding larval development and life-history
evolution in marine invertebrates. Annual Review of Ecology and Systematics
Strathmann, R.R. 2000. Functional design in the evolution of embryos and larvae. Cell
and Developmental Biology 11:395-402.
Tantawy, A.O., and M.R. El-Helw. 1966. Studies on natural populations of Drosophila.
V. Correlated response to selection in Drosophila melanogaster. Genetics 53:97-
Tatar, M., J.R. Carey, and J.W. Vaupel. 1993. Long-term cost of reproduction with and
without accelerated senescence in Callosobruchus maculatus: analysis of age-
specific mortality. Evolution 47:1302-1312.
Thorson, G. 1950. Reproductive and larval ecology of marine bottom invertebrates.
Biological Reviews of the Cambridge Philosophical Society 25:1-45.
Timofeev, S.F., V.V. Sklyar, and M.V. Savinov. 2004. Stabilizing selection on egg size
in the euphausiid, Thysanoessa raschii (M. Sars, 1864) (Euphausiacea) in the
Barents Sea. Crustaceana 77:265-275.
Todd, C.D., and R.W. Doyle. 1981. Reproductive strategies of marine invertebrates: a
settlement timing hypothesis. Marine Ecology Progress Series 4:75-83.
Tucic, N., I. Gliksman, D. Seslija, O. Stojkovic, and D. Milanovic. 1998. Laboratory
evolution of life-history traits in the bean weevil (. klah, i n li de' obtectus): the
effects of selection on developmental time in populations with different previous
history. Evolution 52:1713-1725.
Turner, R.L., and J.M. Lawrence. 1979. Volume and composition of echinoderm eggs:
implications for the use of egg size in life-history models. pp.25-40. In: Stancyk,
S.E. (ed.). 1979. Reproductive ecology of marine invertebrates. University of South
Carolina Press. Columbia, SC.
Unabia, C.R.C., and M.G. Hadfield. 1999. Role of bacteria in larval settlement and
metamorphosis of the polychaete Hydroides elegans. Marine Biology 133:55-64.
Vance, R.R. 1973a. On reproductive strategies in marine benthic invertebrates. American
Vance, R.R. 1973b. More on reproductive strategies in marine benthic invertebrates.
American Naturalist 107:353-361.
VanderWerf, E. 1992. Lack's clutch size hypothesis: an examination of the evidence
using meta-analysis. Ecology 73:1699-1705.
Warren, D.C. 1924. Inheritance of egg size in Drosophila melanogaster. Genetics 9:41-
Warner, R.R. 1984a. Deferred reproduction as a response to sexual selection in a coral
reef fish: a test of the life historical consequences. Evolution 38:148-162.
Warner, R.R. 1984b. Mating behavior and hermaphroditism in coral reef fishes.
American Scientist 72:128-136.
Wendt, D.E. 1996. Effect of larval swimming duration on success of metamorphosis and
size of the ancestrular lophophore in Bugula neritina (Bryozoa). Biological Bulletin
Wendt, D.E. 1998. Effect of larval swimming duration on growth and reproduction of
Bugula neritina (Bryozoa) under field conditions. Biological Bulletin 195:126-135.
Wray, G.A. 1996. Parallel evolution of nonfeeding larvae in echinoids. Sysstematic
Wray, G.A., and R.A. Raff. 1991. The evolution of developmental strategy in marine
invertebrates. Trends in Ecology and Evolution 6:45-50.
Wright, W.G. 1989. Intraspecific density mediates sex-change in the Patellacean limpet
Lottia gigantea. Marine Biology 100:353-364.
Zwaan, B., R. Bijlsma, and R.R. Hoekstra. 1995. Artificial selection for developmental
time in Drosophila melanogaster in relation to the evolution of aging: direct and
correlated responses. Evolution 49:635-648.
I was born in North Carolina and raised in south Florida. I have traveled
extensively as an adult, and settled in Texas where I renewed my pursuit of education and
received an Associate of Arts degree from Texas Southmost College in Brownsville,
Texas and a Bachelor of Science degree, summa cum laude, from Texas A&M
University. I earned my Master of Science degree from Florida Institute of Technology
in 2000 and continued to earn my Ph.D. in 2006 from the University of Florida with a
major in zoology.