224

Estimation and interpretation of egg provisioning in marine

invertebrates
Larry R. McEdward3,* and Benjamin G. Miner1,2,y
of Zoology, University of Florida, Gainesville, FL 32611-8525, USA; yBodega Marine Laboratory,
University of California, Davis, Bodega Bay, CA 94923-0247, USA

*Department

Introduction
The amount of resources that parents invest in an
offspring can influence the evolution of life histories
by affecting both the fitness of parents and offspring
(Roff 1992; Stearns 1992; Havenhand 1995). Among
animals that free-spawn, such as many species of
marine invertebrates, the amount of resources a
mother provisions an egg represents the entire parental
investment (Levin and Bridges 1995). This is because
offspring develop independently in the water column
without any parental care. Thus, changes in parental
investment per offspring can only occur by changing
the amount or type of material placed in an egg. As
a result, per-offspring maternal investment is often
measured as egg size (diameter or volume) or egg
energy for marine invertebrates.
A number of theoretical models have been developed
to understand the selective forces that shape the patterns of per-offspring maternal investment in marine

invertebrates (Vance 1973; Christiansen and Fenchel

1979; Perron and Carrier 1981; Grant 1983;
Roughgarden 1989; McEdward 1997; Levitan 2000;
McEdward and Miner 2003), and these have provided
important insights into the evolution of life histories
(Havenhand 1995). Unfortunately, empirical tests of
these models are hampered by the fact that the parameter these models use to represent per-offspring
maternal investment, referred to as s, is rarely empirically measured. Although egg size has often been
assumed to represent s, McEdward and Janies (1997)
demonstrated that this assumption is incorrect and
that past attempts to test these models by comparing
the predictions of a model with distributions of egg
sizes are misleading.
In order to ameliorate this situation it is necessary to
develop and evaluate ways for empirically estimating s.
Here, we review the definition of s and how to interpret values of s in the light of marine invertebrate

From the symposium Complex Life-Histories in Marine Benthic Invertebrates: A Symposium in Memory of Larry McEdward presented at the
annual meeting of the Society for Integrative and Comparative Biology, January 48, 2005, at San Diego, CA.
1 E-mail: bgminer@ucdavis.edu
2 Present address: Department of Biology, Western Washington University, Bellingham, WA 98225-9160, USA
3 Deceased
Integrative and Comparative Biology, volume 46, number 3, pp. 224232
doi:10.1093/icb/icj026
Advance Access publication March 29, 2006
The Society for Integrative and Comparative Biology 2006. All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org.

Downloaded from http://icb.oxfordjournals.org/ at University of Thessaly on September 20, 2012

Synopsis Per-offspring maternal investment is an integral part of life-history theory. To understand the evolution of
per-offspring maternal investment in marine invertebrates, a number of mathematical models have been developed. These
models examine how selection affects the proportion of maternally derived egg energy used to produce a newly metamorphosed
juvenile (s) and make predictions about the distribution of s in nature. However, there are very few published values of s and
therefore it is difficult to evaluate how well these models match nature. We present several equations to empirically
estimate values of s for any group of marine invertebrate, and use data from echinoderms to compare the different equations.
The calculations that directly estimate s require information on the amount of egg energy, juvenile energy, and energy
metabolized during development. Currently, there are few data available for directly estimating s, and thus generating distributions of s derived from direct estimates is not possible. Furthermore, the direct estimations of s are informative for
planktotrophy but not for lecithotrophy. We have developed an equation that can be used to directly estimate s for lecithotrophs.
The calculations to indirectly estimate s only require egg energy or egg size for the species in question and the value of s and
egg energy or size for a reference species. This reference species replaces the need to measure juvenile energy and energy
metabolized during larval development. Because egg energy or size is currently available for many species, the indirect estimates
will be useful for generating distributions of s, and will allow comparisons with models. Although these indirect methods are
good for generating distributions of s, they do not provide reliable estimates of s for any particular species. Estimating values of s
to compare models is a critical gap in our current evaluations of marine invertebrate life-history models.

225

Egg provisioning in marine invertebrates

developmental modes. We then review equations

investigators have used to calculate s. To evaluate these
equations, we estimate values of s for echinoderms
and discuss advantages and disadvantages of each
equation. The purpose of this review is to provide a
conceptual framework for measuring s in nature for
any group of marine invertebrates, so that theory can
be empirically tested. A better integration of theoretical
and empirical studies is essential to better understand
the evolution of life histories in marine invertebrates.

Vances definition of s

Energy egg
Energy lecithotrophic

1
egg

In other words, s represents the amount of energy in an

egg scaled to the minimum amount of egg energy needed
to produce a newly metamorphosed juvenile without any
exogenous food.
Because egg energy only represents the numerator
of this equation, it is clear that simply measuring
egg energy or estimating egg energy from egg size is
insufficient for calculating s. Both egg energy (the
numerator) and the energy required for metamorphosis without food (the denominator) must be measured
to calculate s.

Values of s and developmental mode

Despite the huge influence of Vances model and its
predictions about per-offspring maternal investment in
terms of s, only recently has the translation between s
and two common modes of development in marine
invertebrates, planktotrophy and lecithotrophy,
been explicitly discussed (Herrera and others 1996;

Emprical estimates of s
Larval energy budget
Hoegh-Guldberg and Emlet (1997) presented the
first direct estimation of s for the sea urchins
Heliocidaris
erythrogramma
and
Heliocidaris
tuberculata, although they did no put their finding
in these terms. They estimated the numerator in equation 1 by measuring the energy of an egg, and the
denominator by measuring (1) the energy used in
basic metabolic functions and developmental processes
and (2) the energy content of the metamorph. With
these three measurements, s can be calculated using the
following equation.
s

Energy egg
Energy metabolism Energy metamorph

Proportional egg energy

If we assume that larval metabolism is a small fraction
of the denominator in equation 2, which maybe the

Downloaded from http://icb.oxfordjournals.org/ at University of Thessaly on September 20, 2012

Vance (1973) developed the first theoretical model to

explain the evolution of maternal investment per offspring for free-spawning benthic marine invertebrates.
This model has been extremely influential and has
provided the theoretical framework for much of larval
ecology (Strathmann 1985; Havenhand 1995). Most
subsequent models of life-history evolution in marine
invertebrates have been developed as modifications
or extensions of the Vance model (Christian and
Fenchel 1979; Perron and Carrier 1981; Grant 1983;
Roughgarden 1989; McEdward 1997; Levitan 2000;
McEdward and Miner 2003). All these models use
the same parameter, s, developed by Vance (1973) to
represent maternal investment per offspring and treat
s as the quantity that natural selection acts on to
maximize fitness.
Vance (1973) defined s, as the egg energy content
measured not in calories per se but in units of energy
sufficient for a lecithotrophic egg (eqn. 1).

McEdward and Janies 1997; Levitan 2000).

Planktotrophy is characterized by insufficient egg
energy to complete larval development, requiring
feeding during the larval stage, whereas lecithotrophy
is defined as an adequate amount of egg energy
to complete larval development without feeding
(McEdward and Janies 1993).
From Vances definition of s, these two developmental modes are easily interpreted in terms of s (Herrera
and others 1996). Planktotrophy is defined as values of
s > 0 and <1 because there is insufficient egg energy to
complete larval development without feeding.
Thus larval feeding is necessary, and the lower the
value of s, the more energy that must be acquired
during the larval phase in order to complete larval
development and metamorphose. In principle if
s 0.999, then larval feeding is still necessary because
there is insufficient energy in the egg to complete larval
development, though 99.9% of development can be
fueled by egg energy.
Although Vance (1973) only considered one value
of s that corresponded to lecithotrophy, s 1 (see
McEdward and Janies 1997), values of s
1 indicate
that eggs have enough energy to complete larval development without exogenous food and are therefore
lecithotrophic. Furthermore, values of s > 1 signify
that egg energy remains after metamorphosis and
the larger the value of s the greater the amount of
remaining egg energy. For example, a lecithotroph
with a value of s 100 requires only 1% of its egg
energy for larval development and has 99% of its
egg energy available for growth and development
after metamorphosis.

226

L. R. McEdward and B. G. Miner

case for some species, like H. erythrogramma (HoeghGuldberg and Emlet 1997), then we can simplify
equation 2, and s can be estimated by measuring
only the egg energy and the energy in the newly metamorphosed juvenile (eqn. 3). This equation may be
preferred because it eliminates the labor and timeconsuming measurements of larval metabolisms.
s

Energy egg
Energy metamorph

Relative egg energy or volume

EggEnergy species


1
EggEnergyreference * Sreference

For many species of marine invertebrates, only egg sizes

are known, and therefore all of the above calculations
(eqns. 24) are not an option. If we are willing to assume
that egg size is a reasonable index of egg energy, egg size
can be substituted for egg energy in equation 4. Studies
have shown that indeed there is a good relationship
between egg energy and size among species (for example,
Jaeckle 1995; McEdward and Morgan 2001), but not
within a species (McEdward and Coulter 1987). To estimate s with only egg size information, three values are
needed: (1) the egg size of the species in question, (2) the
egg size of a reference species, and (3) the value of s for
that reference species. Volume is the most appropriate

EggVolumespecies


1
EggVolumereference * Sreference

Calculations of s for echinoderms

To evaluate these different equations, we calculated s
for species of echinoderm. We have used echinoderms
because the needed information for these equations
(for example, egg size and egg energy) is known
(Emlet and others 1987; Jaeckle 1995; HoeghGuldberg and Emlet 1997; McEdward and Morgan
2001). We were able to calculate s for 2 species with
the larval energy budget equation (eqn. 2), for
4 species with the proportional energy equation
(eqn. 3), and for 39 species with the relative egg energy
and volume equation (eqns. 3 and 4).
With the larval energy budget equation and based on
the data provided from Hoegh-Guldberg and Emlet
(1997), we calculated s for two species of sea urchin.
For H. tuberculata, which has a planktotrophic larva,
s equaled 0.0168, and for H. erythrogramma, which
has a lecithotrophic larva, s equaled 0.9480 (Table 1).
With the proportional energy equation, we calculated s
values for these two species and two additional species.
For the two species with planktotrophic larvae (both
sea urchins), the values were 0.073 for Paracentrotus
lividus and 0.023 for H. tuberculata (Table 1), based on
data from Fenaux and colleagues (1985) and HoeghGuldberg and Emlet (1997). For the two species with
lecithotrophic larvae (a crinoid and a sea urchin), the
values were 5.296 for Florometra serratissima and 1.000
for H. erythrogramma (Table 1), based on data from
McEdward and colleagues (1988) and Hoegh-Guldberg
and Emlet (1997).
With the relative egg energy and volume equations
(eqns. 4 and 5), we calculated s in three different ways:
(1) with egg energy, (2) with egg energy predicted from
egg volume, and (3) with egg volume. For each of these,
we used Clypeaster rosaceus as our reference species.
Egg energy has been measured for C. rosaceus (Miner
and others 2002), and because it is near the border of
lecithotrophy and planktotrophy (Emlet 1986), s for
C. rosaceus is assumed to be 1 (see McEdward
and Janies 1997; Levitan 2000). Because this is only
a rough estimation of s for C. rosaceus, we also used
H. tuberculata as a reference species for calculating s
from egg energy. We chose H. tuberculata because
we had a direct estimate of s from the larval energy
budget calculation.
Based on the data from McEdward and Morgan
(2001), values of s calculated from egg energy for

Downloaded from http://icb.oxfordjournals.org/ at University of Thessaly on September 20, 2012

The larval energy budget (eqn. 2) and proportional

energy (eqn. 3) equations both require timeconsuming methods (for example, larval rearing and
measurements of larval metabolic rates) to estimate s.
As a result, there are only a few species for which we can
currently estimate s with these methods. However,
because one goal is to compare the predictions of
theoretical models with distributions of s values, it is
necessary to empirically estimate s for many species.
Levitan (2000) presented a method to easily estimate
values of s for many species. His method uses a reference species, for which egg energy and the value of s are
known, to estimate the minimum energy needed for a
lecithotrophic egg (that is, the denominator of eqn. 1).
The only additional information that is needed to
calculate s is the egg energy (that is, the numerator)
for a number of species of a particular group. This
eliminates the difficult task of estimating the denominator in the above equations (eqns. 2 and 3). Of
course the assumption is that the reference species is
an accurate estimate of the denominator for each
species for which s is calculated. With this method,
values of s can be calculated for a species if three
measures are known: (1) the egg energy of the species
in question, (2) the egg energy of a reference species,
and (3) the value of s for that reference species.

measure of egg size for estimating the level of provisioning, so s can be calculated as follows.

227

Egg provisioning in marine invertebrates

Table 1 A comparison of s values estimated with equations 25
Species
Paracentrotus lividusa
b

Larval
pattern

s (larval energy
budget)

s (proportional
energy)

sa (relative
energy)

sb (relative
energy)

sc (relative
energy)

s (relative
volume)

pl

0.0737

0.0174

0.0052

0.0146

0.0380

Heliocidaris tuberculata

pl

0.0168

0.0232

0.0202

0.0001

0.0170

0.0416

Florometra serratissimac

lec

5.2964

0.2792

0.7033

0.2346

0.6873

Heliocidaris erythrogrammab

lec

0.9480

1.0000

3.3222

2.6756

2.7921

2.9659

Larval patterns are obligate planktotrophy (pl) and obligate lecithotrophy (lec). sa and sb used Clypeaster rosaceus as the reference
(data from Miner and others 2002), sb used egg volume to estimate energy (see text), and sc used Heliocidaris tuberculata
as the reference (data from Hoegh-Guldberg and Emlet 1997).
aData from Fenaux and colleagues (1985).
bData from Hoegh-Guldberg and Emlet (1997).
cData from McEdward and colleagues (1988).

Evaluation and comparison of the

different estimations of s
We took several general approaches to evaluate the
different equations. First, we checked whether the
calculated values of s differed among the different
calculations. Next, we checked whether the calculated
values of s corresponded to the developmental mode
for each species. In other words, were values of s <1 for
species with planktotrophic larvae and
1 for lecithotrophic species? Lastly, for the relative egg energy and
volume calculations, where we had values of s for a
number of species, we also compared the distribution
of s values derived from the different methods. This
latter approach was intended to alleviate some of
the problems associated with error in the data. Any
comparison between species or methods for a species
is influenced by error in the data used to calculate

values of s. Looking at the distributions of s provided

a way to see a pattern despite any underlying error
(assuming the error is random).
During our evaluations of the different equations, it
became evident that there are problems with some of
the equations. These problems are (1) the inability of
the larval energy budget and proportional energy
equations (eqns. 2 and 3) to accurately estimate s for
lecithotrophs, (2) the inability of the relative egg
energy or volume equation (eqn. 4) to accurately estimate s when the reference species does not accurately
predict the minimum amount of energy for lecithotrophy in that species, and (3) the issue of scaling between
egg energy and egg volume and its effects on the
calculations of s when egg size is used.
The discrepancy between the direct calculations of s
(eqns. 2 and 3) and the relative calculations (eqns. 4
and 5) and the lack of correspondence between the
known developmental mode and the calculated values
of s for two lecithotrophic species highlight the first two
problems. For H. erythrogramma the direct calculations
indicate that s is 1, whereas the relative calculations
indicate that s is 3 (Table 1). Clearly the value of s for
H. erythogramma should be
1 because it has a lecithotrophic larva. In addition, a value of s 3 seems much
more reasonable than a value of 1, because this species
has a nonfeeding larva (measurement error probably
explains why the value derived from the larval energy
budget of 0.948 was <1). Species that are lecithotrophic but close to the boundary between planktotrophy and lecithotrophy typically have facultative feeding
larvae, for example C. rosaceus and Brisaster latifrons
(Emlet 1986; Hart 1996). Thus, the direct equations
(eqns. 2 and 3) appear to be inaccurate for lecithotrophic species.
This underestimation of s from the direct calculations for lecithotrophic species is caused by the value
measured to represent the denominator (that is, the
minimum amount of energy required to produce a
lecithotrophic larva). With the direct calculations the

Downloaded from http://icb.oxfordjournals.org/ at University of Thessaly on September 20, 2012

planktotrophs ranged from 0.008 to 0.284, whereas

for lecithotrophs the values ranged from 0.279 to
84.07 when C. rosaceus was used as the reference
species (Table 2). We obtained similar values when
H. tuberculata was used as the reference species; values
of s for planktotrophs ranged from 0.006 to 0.239 and
those of lecithotrophs ranged from 0.235 to 70.65
(Table 2). We calculated s from estimates of egg energy
derived from egg volume using the allometric
regression equation estimated for echinoderms
by McEdward and Morgan (2001)EggEnergy
0.012 9.644 EggVolume0.868. Other allometric
relationships between egg size and energy should be
used for other taxa (see Pernet and Jaeckle 2004).
Estimated values of s for planktotrophs with C. rosaceus
as the reference species ranged from 0.031 to 0.489,
and those for lecithotrophs ranged from 0.703 to 54.79
(Table 2). Using egg volumes and C. rosaceus as the
reference species, values of s for planktotrophic species
ranged from 0.021 to 0.472 and those for lecithotrophs
ranged from 0.687 to 95.53 (Table 2).

228

L. R. McEdward and B. G. Miner

Table 2 Calculation of s for 38 species of free-spawning echinoderms using relative egg energy and relative egg volume
(eqns.4 and 5)
Larval Diameter Volume Energy
sa (relative sb (relative sc (relative s (relative
energy)
energy)
volume)
pattern (mm)
(ml)
(Joules egg1) energy)

Clypeaster rosaceus

fac

274

0.01080

0.16315

1.0000

1.0000

0.8404

1.0000

Heliocidaris tuberculata

pl

95

0.00045

0.00330

0.0202

0.0001

0.0170

0.0416

Arbacia punctulata

pl

75

0.00022

0.00132

0.0081

0.0311

0.0068

0.0205

Arbacia lixula

pl

77

0.00024

0.00281

0.0172

0.0285

0.0145

0.0222

Strongylocentrotus purpuratus

pl

80

0.00027

0.00165

0.0101

0.0245

0.0085

0.0248

Paracentrotus lividus

pl

92

0.00041

0.00284

0.0174

0.0052

0.0146

0.0380

Aspidodiadema jacobyi

pl

97

0.00049

0.00295

0.0181

0.0052

0.0152

0.0454

Lytechinus variegatus

pl

104

0.00060

0.00528

0.0324

0.0198

0.0272

0.0560

Echinometra lacunter

pl

106

0.00063

0.00224

0.0137

0.0230

0.0115

0.0583

Stylocidaris lineata

pl

110

0.00070

0.00317

0.0194

0.0316

0.0163

0.0648

Coelopleurus floridanus

pl

115

0.00080

0.00784

0.0480

0.0439

0.0404

0.0741

Dendraster excentricus

pl

119

0.00090

0.00328

0.0201

0.0564

0.0169

0.0838

Asterias forbesi

pl

133

0.00124

0.00796

0.0488

0.0951

0.0410

0.1145

Archaeopneustes histrix

pl

134

0.00129

0.00654

0.0401

0.1012

0.0337

0.1195

Strongylocentrotus franciscanus

pl

139

0.00144

0.00577

0.0354

0.1178

0.0297

0.1331

Pisaster ochraceus

pl

154

0.00195

0.00783

0.0480

0.1740

0.0403

0.1806

Strongylocentrotus droebachiensis pl

157

0.00206

0.01218

0.0746

0.1862

0.0627

0.1911

Luidia clathrata

pl

166

0.00245

0.01986

0.1217

0.2266

0.1023

0.2266

Strongylocentrotus pallidus

pl

164

0.00235

0.00904

0.0554

0.2167

0.0465

0.2178

Odontaster validus

pl

169

0.00257

0.01955

0.1198

0.2397

0.1007

0.2382

Parastichopus californicus

pl

189

0.00359

0.00951

0.0583

0.3429

0.0490

0.3326

Encope aberrans

pl

189

0.00359

0.00401

0.0246

0.3430

0.0207

0.3326

Encope michelini

pl

212

0.00510

0.04639

0.2843

0.4891

0.2390

0.4723

Florometra serratissima

lec

241

0.00742

0.04555

0.2792

0.7033

0.2346

0.6873

Cucumaria miniata

lec

493

0.06398

0.82539

5.0592

4.9344

4.2520

5.9253

Acodontaster hodgsoni

lec

547

0.08711

0.97715

5.9894

6.4711

5.0338

8.0677

Psolus chitinoides

lec

569

0.09828

1.04957

6.4333

7.1928

5.4069

9.1018

Echinaster sp. 1

lec

720

0.19912

2.88073

17.6574

13.3332

14.8401

18.4407

Echinaster sp. 2

lec

765

0.23916

4.69026

28.7489

15.6434

24.1619

22.1490

Solaster endeca

lec

811

0.28510

3.55631

21.7984

18.2319

18.3204

26.4034

Echinaster spinulosus

lec

834

0.31000

3.51600

21.5513

19.6114

18.1127

28.7094

Solaster dawsoni

lec

887

0.37250

4.00467

24.5466

23.0125

20.6301

34.4976

Solaster stimpsoni

lec

912

0.40600

4.52640

27.7445

24.8039

23.3178

37.6001

Psilaster charcoti

lec

944

0.44892

2.57138

15.7613

27.0708

13.2465

41.5750

Mediaster aequalis

lec

951

0.45990

5.78792

35.4770

27.6460

29.8166

42.5918

Pteraster tesselatus

lec

1176

0.87000

8.26919

50.6859

48.1277

42.5988

80.5716

Pteraster militaris

lec

1192

0.90478 10.20000

62.5208

49.7956

52.5455

83.7925

Perknaster fuscus

lec

1192

0.90478

3.19897

19.6081

49.7956

16.4796

83.7925

Henricia leviuscula

lec

1236

1.01000 13.71574

84.0705

54.7920

70.6568

93.5372

Larval patterns are obligate planktotrophy (pl), facultative planktotrophy (fac), and obligate lecithotrophy (lec). Egg diameters,
volumes, and energy contents are from McEdward and Morgan (2001). See Table 1 for explanations for the three relative
energy calculations (sa, sb, sc).

Downloaded from http://icb.oxfordjournals.org/ at University of Thessaly on September 20, 2012

Species

229

Egg provisioning in marine invertebrates

Energy egg
Energy egg  Energy juv

6
metabolism

So how do we measure the amount of energy used by the

juvenile for metabolic processes? After larvae metamorphose, the juveniles can be held without food and
the amount of energy in a juvenile can be quantified just
before it starves. Either this value can be subtracted
from the amount of energy in the newly metamorphosed
juvenile to give the amount of energy used for juvenile
metabolic processes, or it can represent the entire denominator of equation 6. Unfortunately this method is likely
difficult and time consuming. In addition, this equation
will overestimate values of s if a bigger juvenile skeleton or
other energetic costs are required to incorporate extra
energy for juvenile growth. Currently there are no data
available to estimate s using this juvenile energy budget
equation (eqn. 6).

Blastomere separation might provide an alternative

way to estimate s for lecithotrophs. By halving or
quartering embryos and rearing the reduced embryos
to metamorphosis, the minimum size required to
produce a juvenile could be estimated and used as
the denominator in equation 2. However, this
method is only viable for a few taxa (for example,
echinoderms) and is also likely time consuming and
labor intensive.
The species F. serratissima highlights the second
problem. Although F. serratissima has a lecithotrophic
larva, the values of s calculated with the relative equations were <1 and are clearly incorrect. These low s
values of F. serratissima are because this species has a
very small egg but also a very small juvenile, and as a
result it can still be lecithotrophic despite the small egg
size (McEdward and others 1988). Thus, the small
value of egg energy or volume is divided by a reference
species that requires more energy to produce a juvenile,
and the value of s is underestimated. Although rather
obvious, this highlights that a particular reference species is not suitable for all species. Investigators should
acknowledge this potential problem when calculating s
with the relative equations, and justify why a reference
species is appropriate for that group. Alternatively, a
number of appropriate reference species could be used,
allowing the investigator to report a measure of error or
confidence in the measurements.
The calculations from F. serratissima also indicate
that for some species the proportional energy
equation (eqn. 3) will greatly overestimate the value
of s. As mentioned above, the problems with the
direct calculations (eqns. 2 and 3) result in values of
s that are 1 for lecithotrophs. However, the value of
s calculated with the proportional energy equation
for F. serratissima was 5 (Table 1). The reason for this
discrepancy is that its larva uses a large amount of
energy for metabolic processes. Thus the assumption
that larval metabolic costs are negligible, which is
necessary to use the proportional energy equation,
is violated.
The distributions of s values for planktotrophic and
lecithotrophic species show that values of s derived
from egg volume and egg energy can differ. We
observed rather different values of s calculated with
the relative egg energy (eqn. 4) and relative egg
volume (eqn. 5) (Fig. 1). This difference is expected
whenever the relationship between egg energy and egg
volume is nonlinear, and is likely the explanation for
the differences we observed (McEdward and Morgan
2001). In addition, for five planktotrophic species, we
obtained negative values of s with the relative egg
energy equation (eqn. 4) when we estimated egg
energy from egg size (see sb in Table 2). Negative values

Downloaded from http://icb.oxfordjournals.org/ at University of Thessaly on September 20, 2012

denominator is estimated by measuring the energy of

the juvenile. However, if a species provisions its eggs
with more energy than the minimum needed to be
lecithotrophic, then this energy will be present in the
juvenile, and the denominator is overestimated. In
other words, the more energy a lecithotrophic species
provisions its egg, the larger the values of both the
numerator and denominator, and therefore the
value of s will remain approximately the same. Thus,
with the direct equations (eqns. 2 and 3) it is likely that
the value of s will be underestimated to some degree for
all lecithotrophic species because all lecithotrophic
species, even facultative planktotrophs, are provisioned
with more energy than needed to just metamorphose.
This problem raises another potential problem when
calculating s with the direct equations for species
with feeding larvae (either planktotrophic or lecithotrophic). If larvae consume more food than the minimum amount necessary to metamorphose then the
denominator will be overestimated and the value
of s underestimated. Despite the potential for this
problem, we have no evidence to suggest that it is a
problem. For both species with feeding larvae in
Table 1, the direct calculations of s were greater (not
less than) than the indirect calculations.
Because the direct equations (eqns. 2 and 3) are
uninformative for lecithotrophic species, we need an
alternative way to directly calculate s for lecithotrophs. A possible solution is to calculate the energy
needed to build the juvenile for lecithotrophs. By estimating how much energy is used after metamorphosis,
we can subtract this amount from the egg energy
and estimate the amount of energy required to just
produce a juvenile. We might therefore consider using
the following equation to estimate s for lecithotrophic
species.

230

L. R. McEdward and B. G. Miner

Planktotrophic Species
Lecithotrophic Species
Relative Energy (sa)
3

17.5
15
12.5
10
7.5
5
2.5

2.5
2
1.5
1
0.5
0

0.2

0.4

0.6

0.8

20

60

80

40

60

80

40

60

80

40

60

80

3
2.5
2
1.5
1
0.5
0

0.2

0.4

0.6

0.8

20

Relative Energy (sc)

4

17.5
15
12.5
10
7.5
5
2.5

3
2
1
0

0.2

0.4

0.6

0.8

20

Relative Volume (s)

10

1.5

6
1
4
0.5

2
0

0.2

0.4

0.6

0.8

20

Value of s
Fig. 1 Distributions of s for planktotrophic and lecithotrophic species. See Table 1 for explanations for the three relative
energy calculations (sa, sb, sc).

of s are not biologically possible and indicate that deriving egg energy from egg volume can produce inaccurate estimates. This is likely because the confidence
intervals of regressions are poorest at the extremes.
Lastly, values of s calculated with the relative egg
energy equation (eqn. 4) were very similar even when
we used different reference species (Fig. 1).

Conclusions
In this article, we have presented several equations
to empirically estimate s for marine invertebrates

and have discussed their advantages and limitations.

The relative egg energy and relative egg volume
equations (eqns. 4 and 5) allow researchers to estimate
values of s for a number of species and are most appropriate for generating data to test theoretical models.
However, because these equations require a reference
species, they cannot take into account changes in
the cost of development and production of the juvenile
(that is, the denominator of equation 1). Therefore
any estimate of s for a single species is probably not
very reliable and should be considered with caution
until further experimental work determines how

Downloaded from http://icb.oxfordjournals.org/ at University of Thessaly on September 20, 2012

Frequency

Relative Energy
7
6
5
4
3
2
1

40

(sb)

Egg provisioning in marine invertebrates

Acknowledgments
A. O. D. Willows, Director, provided space and facilities at the Friday Harbor Laboratories, University of
Washington. We thank the following people for helpful
discussions of the ideas presented in this article or
for comments on the manuscript: J. Cowart,
M. Hart, D. Levitan, A. Moran, D. Padilla,
J. Pechenik, R. Strathmann, C. Osenberg, C. St
Mary, and E. Werner. Support was provided
by National Science Foundation grants OCE
9819593 to L.R.McE. and OCE 0385028 to
S. Morgan and B.G.M. (Contribution 2295 Bodega
Marine Laboratory, University of California, Davis).

References
Christiansen FB, Fenchel TM. 1979. Evolution of marine invertebrate reproductive patterns. Theor Popul Biol 16:26782.

Emlet RB. 1986. Facultative planktotrophy in the tropical

echinoid Clypeaster rosaceus (Linnaeus) and a comparison
with obligate planktotrophy in Clypeaster subdepressus
(Gray) (Clypeasteroida, Echinoidea). J Exp Mar Biol Ecol
95:183202.
Emlet RB, McEdward LR, Strathmann RR. 1987. Echinoderm
larval ecology viewed from the egg. In: Jangoux M,
Lawrence JM, editors. Echinoderm studies. Vol. 2, Lisse,
Netherlands: A.A. Balkema.:55136.
Fexaux L, Cellario C, Etienne M. 1985. Larval growth of the sea
urchin Paracentrotus lividus. Mar Biol 86:1517.
Grant A. 1983. On the evolution of brood protection in marine
benthic invertebrates. Am Nat 122:54955.
Hart MW. 1996. Evolutionary loss of larval feeding: development, form and function in a facultatively feeding larva,
Brisaster latifrons. Evolution 50:17487.
Havenhand JN. 1995. Evolutionary ecology of larval types. In:
McEdward LR, editor. Ecology of marine invertebrate larvae.
Boca Raton, FL: CRC Press. p 79122.
Herrera JC, McWeeney SK, McEdward LR. 1996. Diversity of
energetic strategies among echinoid larvae and the transition
from feeding to nonfeeding development. Oceanologica Acta
19:31321.
Hoegh-Guldberg O, Emlet RB. 1997. Energy use during
the development of a lecithotrophic and a planktotrophic
echinoid. Biol Bull 192:2740.
Jaeckle WB. 1995. Variation in the size, energy content, and
biochemical composition of invertebrate eggs: correlates to
the mode of larval development. In: McEdward LR, editor.
Ecology of marine invertebrate larvae. Boca Raton, FL: CRC
Press. p 4978.
Levin LA, Bridges TS. 1995. Pattern and diversity in reproduction and development. In: McEdward LR, editor. Ecology of
marine invertebrate larvae. Boca Raton, FL: CRC Press.
p 148.
Levitan DR. 2000. Optimal egg size in marine invertebrates:
Theory and phylogenetic analysis of the critical relationship between egg size and development time in echinoids.
Am Nat 156:17592.
McEdward LR. 1997. Reproductive strategies of marine benthic
invertebrates revisited: Facultative feeding by planktotrophic
larvae. Am Nat 150:4872.
McEdward LR, Coulter LK. 1987. Egg volume and energetic
content are not correlated among sibling offspring of
starfish: implications for life-history theory. Evolution
41:9147.
McEdward LR, Janies DA. 1993. Life cycle evolution in asteroids:
what is a larva? Biol Bull 184:25568.
McEdward LR, Janies DA. 1997. Relationships among
development, ecology, and morphology in the evolution of
Echinoderm larvae and life cycles. Biol J Linn Soc Lond
60:381400.
McEdward LR, Morgan KH. 2001. Interspecific relationships
between egg size and the level of parental investment per
offspring in echinoderms. Biol Bull 200:3350.

Downloaded from http://icb.oxfordjournals.org/ at University of Thessaly on September 20, 2012

different juvenile size and larval metabolic costs are

among species. The larval energy budget and
proportional energy calculations (eqns. 2 and 3)
provide a more direct way of quantifying s and overcome the problems with the relative egg energy and
relative egg volume equations. However, it unclear
whether the proportional energy calculation can provide reliable estimates because it is currently unknown
whether we can assume that energy spent on larval
metabolism is sufficiently small and can therefore be
ignored. The larval energy budget calculation should
provide the most accurate and reliable values of s.
However, it is very time consuming and likely unreliable for lecithotrophs. The juvenile energy budget
equation (eqn. 6) is one solution for directly estimating
s for lecithotrophs. Despite the time needed to obtain
the data for these direct estimates (eqns. 2 and 6), they
are important because they provide the most accurate
estimates of s. Furthermore, they can provide data
needed for the relative estimations (eqns. 4 and 5).
The direct estimates (eqns. 2 and 6) can also be used
to verify the relative estimates (eqns. 4 and 5), and if
contradictions arise they provide information on the
causes for different estimationsfor example are costs
of development and production of the juvenile different than the reference species (that is, the denominator
of eqn. 1).
Empirical calculation and interpretation of s
represents a gap in our understanding of the life
histories of marine invertebrates, which is necessary
to understand the evolution of reproductive strategies
in marine invertebrates. Our critical evaluation of
different ways to empirically estimate s should serve
as a guide for future investigations and allow more
accurate comparisons between models and nature.

231

232
McEdward LR, Miner BG. 2003. Fecundity-time models
of reproductive strategies in marine benthic invertebrates:
fitness differences under fluctuating environmental conditions. Mar Ecol Prog Ser 256:11121.
McEdward LR, Carson SF, Chia FS. 1988. Energetic content of
eggs, larvae, and juveniles of Florometra serratissima and the
implications for the evolution of crinoid life histories. Int J
Invert Reprod Dev 13:922.
Miner BG, Cowart JD, McEdward LR. 2002. Egg energy for the
facultative planktotroph Clypeaster rosaceus (Echinodermata:
Echinoidea), revisited. Biol Bull 202:979.

Perron FE, Carrier RH. 1981. Egg size distributions among

closely related marine invertebrate species: are they bimodal
or unimodal? Am Nat 118:74955.
Roff DA. 1992. The evolution of life histories: Theory and analysis. New York, NY: Chapman & Hall.
Roughgarden J. 1989. The evolution of marine life cycles. In:
Feldmans FMW, editor. Mathematical evolutionary theory.
Princeton, NJ: Princeton University Press. p 270300.
Stearns SC. 1992. The evolution of life histories. New York, NY:
Oxford University Press.
Strathmann RR. 1985. Feeding and nonfeeding larval development and life-history evolution in marine invertebrates. Ann
Rev Ecol Syst 16:33961.
Vance RR. 1973. On reproductive strategies in marine benthic
invertebrates. Am Nat 107:33952.

Downloaded from http://icb.oxfordjournals.org/ at University of Thessaly on September 20, 2012

Pernet B, Jaeckle WB. 2004. Size and organic content of eggs

of marine annelids, and the underestimation of egg energy
content by dichromate oxidation. Biol Bull 207:6771.

L. R. McEdward and B. G. Miner