Ecology, 86(6), 2005, pp. 1571–1579
q 2005 by the Ecological Society of America
LATITUDINAL VARIATION IN PALATABILITY OF SALT-MARSH PLANTS:
ARE DIFFERENCES CONSTITUTIVE?
CRISTIANO S. SALGADO
AND
STEVEN C. PENNINGS1
Department of Biology and Biochemistry, University of Houston, Houston, Texas 77204 USA
Abstract. Biogeographic theory argues that consumer–prey interactions are more intense, and prey defenses better developed, at lower latitudes. Along the Atlantic Coast of
the United States, low-latitude salt marsh plants are less palatable than high-latitude conspecifics. To test the hypothesis that latitudinal variation in palatability would occur in the
absence of geographically different environmental cues (i.e., that differences in palatability
are constitutive rather than induced by climate or herbivore damage), we grew high- and
low-latitude individuals of three species of salt marsh plants from seeds ( Solidago sempervirens) or rhizome cuttings (Distichlis spicata and Spartina alterniflora) in a commongarden greenhouse environment, and compared their palatability to herbivores over time.
We also quantified leaf toughness and nitrogen content over time in order to help explain
results of feeding assays. High-latitude plants were always more palatable to herbivores
than low-latitude conspecifics. Latitudinal variation in plant traits depended on the plant
species. Toughness varied as a function of latitude for Spartina, with low-latitude plants
being consistently tougher than high-latitude conspecifics. For all generations of Spartina,
and for seed-propagated Solidago, high-latitude plants had a higher nitrogen content than
low-latitude conspecifics. The fact that latitudinal differences in palatability and traits of
salt marsh plants persisted in a common-garden environment suggests that this variation
is constitutive, and likely under genetic control, rather than a plastic response to environmental cues. These results are consistent with the theory that latitudinal differences in
herbivory have selected for geographical variation in plant palatability, although we cannot
rule out other selective forces that may also vary across latitude.
Key words: biogeography; common-garden experiment; herbivory; latitude; plant–herbivore interactions; salt marsh.
INTRODUCTION
A central paradigm of biogeography is that consumer–prey interactions vary across latitude, with predation and herbivory being more intense in the tropics
than the temperate zones. Theory predicts that increased consumer pressure at low vs. high latitudes
should select for increased defenses of prey (McArthur
1972, Bakus and Green 1974, Vermeij 1978, Jeanne
1979, Coley and Aide 1991, Jablonski 1993). In the
case of plant–herbivore interactions, greater herbivore
pressure will increase the risk of damage and the potential benefit of defenses, favoring enhanced defenses
against herbivory (Feeny 1976, Levin 1976, Coley et
al. 1985). Herbivore pressure might be greater at low
latitudes because of less severe abiotic control of herbivore populations or greater predictability and productivity of plants (Coley and Aide 1991).
Most studies on latitudinal variation in plant–herbivore interactions are consistent with the general hypothesis that tropical plants are better defended than
temperate ones (Hay and Fenical 1988, Coley and Aide
Manuscript received 12 August 2004; revised 1 November
2004; accepted 2 November 2004. Corresponding Editor: J. T.
Cronin.
1 Corresponding author. E-mail: [email protected]
1991, Bolser and Hay 1996, Pennings et al. 2001),
although brown algae and boreal trees show no trend
or the opposite trend (Van Alstyne and Paul 1990,
Steinberg 1992, Steinberg et al. 1995, Swihart et al.
1994). The most comprehensive data set on this topic
comes from work in Atlantic Coast salt marshes. Palatability trials with fresh plants conducted in multiple
seasons and years demonstrated that both high- and
low-latitude herbivores strongly preferred to eat highvs. low-latitude plants of 10 species (the bulk of the
plant community) (Pennings et al. 2001). Several plant
traits, including toughness, chemical defenses, and N
content also varied across latitude and may be the proximate factors explaining this striking geographic pattern in palatability (Siska et al. 2002). Although we
know quite a bit about latitudinal variation in plant
palatability in this system, and have an emerging understanding of how plant traits contribute to variation
in palatability, work to date has not examined whether
latitudinal variation in palatability is constitutive or is
a plastic response dictated by differences in the environmental conditions experienced by the plants.
Herbivore pressure can select for either constitutively expressed or inducible plant defenses (Karban
and Baldwin 1997, Agrawal 1998). Plants are highly
plastic for a variety of traits, and individuals within a
1571
1572
CRISTIANO S. SALGADO AND STEVEN C. PENNINGS
species may largely vary in growth rate, size, reproduction, allocation to different organs, and chemical
composition (Callaway et al. 2003). Furthermore, phenotypic plasticity in chemical defense is thought to play
a major role in plant–herbivore interactions, with cues
associated with herbivores causing an increase or induction of the defensive phenotype of the plant (Agrawal et al. 2002). Thus, latitudinal differences in plant
palatability might be constitutive or might be induced
by latitudinal differences in environmental cues such
as herbivore damage or physical stress. We know almost nothing about the importance of these two mechanisms in producing geographic patterns in plant palatability.
Here, we test the hypothesis that latitudinal variation
in palatability of salt marsh plants will occur in the
absence of geographically different environmental cues
(i.e., that differences in palatability are constitutive
rather than induced by climate or herbivore damage).
To test this hypothesis, we grew high- and low-latitude
individuals of three species of salt marsh plants from
seeds or rhizome cuttings in a common-garden greenhouse environment, and compared their palatability to
herbivores over time.
METHODS
Collection sites
Salt marshes are the dominant intertidal habitat along
the Atlantic Coast of the United States (Pennings and
Bertness 2001), and offer an attractive system for examining latitudinal variation in plant palatability. Similar plant and animal communities occur across a wide
latitudinal range, making it possible to make intraspecific comparisons across latitude (Pennings and Bertness 1999). Moreover, herbivory in salt marshes can
control the biomass and/or distribution of plants (Pennings and Bertness 2001). To examine latitudinal variation in plant palatability, we worked at sites located
at high latitudes (northeast coast of the United States)
and low latitudes (southeast coast of the United States).
To ensure that working at any one site did not bias the
results, we collected plants and seeds from ten replicate
sites in each region (Appendix A). Logistical constraints precluded sampling of mid-Atlantic or Canadian sites.
Plants and herbivores
We worked with three plant species common at both
high- and low-latitude sites: two C4 grasses, the smooth
cordgrass Spartina alterniflora Loisel and the spikegrass Distichlis spicata (L.) Greene, and a C3 forb, the
seaside goldenrod Solidago sempervirens L. (hereafter
Spartina, Distichlis, and Solidago). These species were
selected from the 10 studied by Pennings et al. (2001)
because they were present at all our sites and easy to
propagate from seed or cuttings. They included the
most abundant Atlantic Coast salt marsh plant ( Spar-
Ecology, Vol. 86, No. 6
tina), and representative species of the three major elevation zones (low-marsh, intermediate, and terrestrial
border, respectively). Logistical constraints limited us
to working with only three species.
Solidago plants were propagated by seed. We collected multiple seeds from at least six individual plants
from each of the 20 sites in the fall of 2002. Seeds
from within a site were pooled, brought to the laboratory, and kept refrigerated for five months. Solidago
seeds were germinated in the spring of 2003, and four
to 10 seedlings raised to adulthood per site. Distichlis
and Spartina were propagated clonally. Four individual
ramets of Spartina and four rhizome cuttings, each with
several shoots, of Distichlis were collected from each
site in June of 2002. To minimize the possibility of
collecting multiple individuals of the same genotype,
we collected ramets that were spaced widely from one
another. Genetic diversity in Spartina populations is
high, and thus there is little likelihood of repeatedly
sampling the same clone with collections taken .1 m
apart (Richards et al. 2004). Individual plants were
potted and kept under common-garden conditions and
free of herbivory in a greenhouse at the University of
Houston. All three plant species were potted in a mixture of 60% potting soil (Fafard’s Professional Formula
52 Mix, Fafard, Agawam, Massachusetts, USA) and
40% sand, kept under shadecloth in summer months,
cooled with swamp coolers, and watered heavily with
fresh water. These conditions did not mimic those of
any parental field population but rather provided common conditions favorable for growth of plants from all
sites. Individual ramets of Spartina and Distichlis collected in the field (here referred to as ‘‘generation 1’’)
were allowed to acclimate in the greenhouse for one
month before being used in feeding trials. Successive
clonal generations of Distichlis and Spartina were separated from the mother plant and repotted periodically
as growth rates permitted (approximately two to three
times per growing season). We followed Distichlis for
a total of four generations and Spartina for five generations. Given that we raised Solidago from seed, we
felt that one full generation was sufficient to test our
hypothesis, because maternal effects affecting palatability to herbivores tend to be short-lived (Agrawal
2002).
To compare the relative palatability of plants from
high- and low-latitude sites, we offered them to adult
individuals of the grasshoppers Orchelimum fidicinum
Rehn and Hebard 1907 (Tettigoniidae) (Spartina), Orphulella pelidna Burmeister 1838 (Acrididae) (Distichlis, Spartina), Paroxya clavuliger Serville 1838
(Acrididae) (Solidago), and Dicromorpha elegans
Morse 1896 (Acrididae) (Distichlis). These herbivores
were chosen because they were among the most common leaf-chewing herbivores feeding on each plant
species (C. Salgado and S. Pennings, personal observations). Previous results indicated that the source of
the herbivore (high- versus low-latitude sites) did not
June 2005
LATITUDINAL VARIATION IN PALATABILITY
1573
PLATE 1. Paired feeding assays. From left to right: grasshopper housed individually with a choice of an undamaged leaf
from a high- and a low-latitude Solidago plant; high- and low-latitude leaves of Soldago, Distichlis, and Spartina after being
offered to herbivores in paired-choice feeding assays. In each case, low-latitude leaves (indicated with ‘‘S’’ or a black mark)
received less damage than high-latitude leaves.
affect feeding preferences (Pennings et al. 2001). The
animals used in these feeding trials were collected from
salt marshes around Sapelo Island, Georgia, except the
O. pelidna that were used for the feeding trials with
the fourth clonal generation of Distichlis and the fifth
clonal generation of Spartina, which, for logistical reasons, were collected in salt marshes near Surfside,
Texas.
Feeding trials
We measured relative plant palatability using pairedchoice feeding assays (see Plate 1, Appendix B). Feeding assays were conducted following Pennings et al.
(2001). Consumers were collected in the field, brought
to the laboratory, and experiments were set up approximately one hour after plant material was obtained.
Grasshoppers were housed individually within 1000mL glass jars, and were offered a choice between an
undamaged leaf from a high-latitude plant and one from
a conspecific low-latitude plant. To reduce any effect
of leaf age on herbivore preference, we collected leaves
in standard positions (e.g., second fully expanded leaf).
High- and low-latitude sites were randomly paired,
with clones selected haphazardly from within each site,
and each feeding trial began with up to 23 replicates,
depending on the availability of herbivores. Individual
replicates were checked twice a day and were terminated when substantial feeding (;30%) on at least one
leaf had occurred or at the end of the third day. Replicates in which both leaves were completely consumed
or in which neither leaf was eaten after a period of 72
hours provided no information on latitudinal variation
of plant palatability and were omitted (total of nine
replicates). Individual consumers were used only once.
Feeding trials were conducted for all the clonal generations of Distichlis and Spartina (four and five generations, respectively) and for seed-germinated Soli-
dago that had been raised to adulthood. Feeding was
measured as square millimeters of leaf area consumed.
Consumption was compared between geographic regions with paired t tests. To determine whether the
strength of the preference shown by consumers
changed among clonal generations, we subtracted the
amount consumed of the low-latitude plant from the
amount consumed of the high-latitude plant for each
replicate. Differences in consumption were compared
across generations with ANOVA.
Field experiment
We also compared the palatability of high- and lowlatitude Solidago using potted plants and a natural assemblage of herbivores in the field. Two undamaged
leaves on each of four healthy plants per site (eight
high- and eight low-latitude sites, for a total of 64
plants) were measured and tagged. Potted plants were
placed in the field at the Airport Marsh, on Sapelo
Island, Georgia. Common grasshoppers observed at the
site included Orchelimum fidicinum, Paroxya clavuliger, and Orphulella pelidna) (C. Salgado and S. Pennings, personal observation). Plants were placed in the
high marsh adjacent to naturally occurring Solidago.
Mammalian herbivores were excluded with a fence.
After 15 days, the amount of accumulated damage on
tagged leaves was measured. Values for individual
leaves and plants were averaged to yield a single value
per site. We examined variation in the amount of leaf
tissue consumed between high- and low-latitude source
populations with a two-sample t test, using sites as
replicates. Similar field assays were not conducted with
the other plant species for logistical reasons.
Toughness and nitrogen content measurements
To help explain results of feeding assays, we quantified leaf toughness and nitrogen content, which are
1574
CRISTIANO S. SALGADO AND STEVEN C. PENNINGS
Ecology, Vol. 86, No. 6
We tested four individual leaves per generation per site
from each plant species. Values of individual leaves
were averaged to yield a single value per species per
generation per site. We examined variation in toughness using two-sample t tests.
To measure N content, plant material was freezedried and pulverized using an amalgamator. CHN analysis was performed in the laboratory of Dr. Samantha
Joye, at the University of Georgia, using a ThermoFinnigan Flash Elemental Analyzer (Thermo Electron
Corporation, Waltham, Massachusetts, USA). Nitrogen
content (percentage of dry mass) was arcsine squareroot transformed and compared for each species using
a two-sample t test (two leaves nested within a site and
sites as units of replication).
RESULTS
Feeding trials
FIG. 1. (A) Consumption of high- vs. low-latitude Spartina alterniflora plants in paired feeding trials. Herbivore species used are shown below paired bars. (B) Leaf toughness
of Spartina alterniflora. (C) Leaf nitrogen content of Spartina
alterniflora. Bars represent means 1 SE. Sample sizes and P
values are given above bars. Generation 1 refers to ramets
collected in the field.
important aspects of plant palatability (Pennings et al.
1998). We measured leaf toughness using a penetrometer, similar to that described in Pennings et al. (1998).
Briefly, penetrometer measurements determined the
force necessary to drive a steel rod through leaf tissue.
Despite being raised in a common garden, greenhouse environment, high- and low-latitude plants did
not converge in palatability. Overall, 13 of 13 trials
resulted in a significant preference for high- vs. lowlatitude plants. In most cases, preferences were strong,
with herbivores eating several times more of the highlatitude plant than of the low-latitude plant. This pattern held true across all plant species and generations,
independent of the herbivore used (Spartina, Fig. 1A;
Distichlis, Fig. 2A and B; Solidago, Fig. 3A).
The strength of the preference (difference in consumption of high- vs. low-latitude plants) did not vary
among generations of Spartina (generations 1–4 [herbivore, Orchelimum], ANOVA, F3,70 5 1.10, P 5 0.35;
conclusions do not change if generation 5 [herbivore,
Orphulella] is included). For Distichlis, the strength of
feeding preferences did not vary among generations
for the herbivore Orchelimum (generations 1–3, ANOVA, F2,50 5 0.26, P 5 0.78, conclusions do not change
if generation 4 [herbivore, Orphulella] is included), but
did for the herbivore Dichromorpha, because there was
a stronger difference in consumption within the third
generation than within the first two (generations 1–3,
ANOVA, F2,43 5 8.34, P 5 0.001, conclusions do not
change if generation 4 [herbivore, Orphulella] is included).
For Solidago, herbivores preferred plants germinated
from seeds collected at high- vs. low-latitude sites, regardless of whether assays were conducted in the laboratory or in the field (Fig. 3A). Although we did not
conduct assays with ‘‘wild’’ Solidago collected directly
from the field, previous work by Pennings et al. (2001)
with field-collected plants showed strong preferences
for high- over low-latitude plants (significant in 11 of
17 assays).
Toughness measurements
Greenhouse-raised Spartina differed in toughness
across latitude, but Distichlis and Solidago did not. For
June 2005
LATITUDINAL VARIATION IN PALATABILITY
1575
Spartina, low-latitude plants were tougher than highlatitude conspecifics in all clonal generations except
for the first, which was marginally significant (Fig. 1B).
For Distichlis, low-latitude plants always tended to be
tougher than high-latitude conspecifics, but the differences were small and never significant (Fig. 2C). Similarly, Solidago germinated from seeds collected at
high- vs. low-latitude sites did not differ in toughness
(Fig. 3B). Note that because the penetrometer apparatus
that we used differed from that used by Pennings et al.
(1998), the values are not directly comparable.
Nitrogen content
Greenhouse-raised Spartina and Solidago differed in
nitrogen content across latitude, but Distichlis did not.
High-latitude Spartina plants had a higher N content
than low-latitude conspecifics in all clonal generations
except for the first, which was only marginally significant (Fig. 1C). Nitrogen content of greenhouse-raised
Distichlis did not vary across latitude (Fig. 2D). Solidago germinated from seeds collected at high-latitude
sites had a higher N content than did low-latitude conspecifics (Fig. 3C).
DISCUSSION
Previous work has documented a striking, community-wide pattern of latitudinal variation in palatability
of salt marsh plants along the Atlantic Coast of the
United States (Pennings et al. 2001). Our results presented here indicate that this variation in palatability
is constitutive rather than plastic, and also shed light
on the proximate plant traits mediating this variation.
To the best of our knowledge, this study is the first to
examine the role of fixed vs. plastic mechanisms controlling variation in plant palatability at the latitudinal
scale. The ultimate factors driving latitudinal variation
in plant palatability in coastal salt marshes, however,
remain to be explored. Below, we discuss each of these
points in turn.
We found that high-latitude salt marsh plants were
always more palatable than low-latitude conspecifics
under common-garden conditions after plants were either germinated and raised to adulthood (Solidago) or
propagated through several clonal generations (Spartina and Distichlis). Overall, 13 of 13 feeding-preference trials with four different herbivore species resulted in a significant preference for high- vs. lowlatitude plants. Our results were more consistent than
the tests with field-collected plants reported in Pennings et al. (2001), which always included some nonsignificant trials.
←
FIG. 2. (A and B) Consumption of high- vs. low-latitude
Distichlis spicata plants in paired feeding trials. Herbivore
species used are shown below paired bars. (C) Leaf toughness
of Distichlis spicata. (D) Leaf nitrogen content of Distichlis
spicata. Bars represent means 1 SE. Sample sizes and P values are given above bars. Generation 1 refers to ramets collected in the field.
1576
CRISTIANO S. SALGADO AND STEVEN C. PENNINGS
FIG. 3. (A) Consumption of high- vs. low-latitude Solidago sempervirens plants in paired feeding trials and in the
field. Herbivore species used are indicated below paired bars.
All plants were from generation 2 (plants grown from seeds
that were collected from field [generation 1] plants). (B) Leaf
toughness of Solidago sempervirens. (C) Leaf nitrogen content of Solidago sempervirens. Bars represent means 1 SE.
Sample sizes and P values are given above bars.
The preference for high- vs. low-latitude plants was
similar across different experiments and clonal generations. In the case of seed-germinated Solidago, similar preference patterns were obtained in both laboratory and field experiments. In the case of Spartina, the
direction and strength of the preference shown by consumers did not change over successive clonal generations. In the case of Distichlis, preferences again did
not change over clonal generations when Orchelimum
was used as the herbivore. When Dicromorpha was
used as the herbivore for feeding trials with Distichlis,
Ecology, Vol. 86, No. 6
the preference for northern plants shown in generation
3 was even stronger than the previous generations. Because this pattern occurred with only one of the two
herbivores, it is unlikely to reflect any important change
in plant characteristics, but rather probably reflects
some minor variation in the assay process. In any case,
because the preference increased under common garden
conditions, rather than decreased, it argues all the more
against latitudinal differences being a plastic response
to the environment that would diminish under commongarden conditions. In sum, the fact that latitudinal variation in palatability was maintained under common
garden conditions suggests that latitudinal differences
in palatability of the three plant species tested are not
primarily plastic responses driven by latitudinal variation in environmental conditions, but rather are constitutive differences that may be under fixed genetic
control.
Our experiments focused on the large differences in
palatability that are observed across latitude, and asked
whether these differences would disappear in a common-garden environment. Thus, we did not explicitly
test for the existence of inducible defenses, and we
would be surprised if these did not exist in salt marsh
plants. Rather, our results demonstrate that inducible
defenses, if they exist, are not sufficient to explain the
large differences in palatability observed at the geographic scale.
Theory suggests that constitutive strategies should
be favored when the environment is relatively predictable, whereas induced strategies should be favored
when the environment is unpredictable (Karban and
Baldwin 1997). Salt marsh plants are abundant and in
most cases perennial, and occur in predictable habitats
in low-diversity communities, and therefore should be
highly apparent to herbivores. Moreover, variation in
herbivore pressure is predictable at the geographic
scale (S. C. Pennings, C. S. Salgado, and C. K. Ho,
unpublished data). Thus, our results are consistent with
the notion that constitutive strategies are favored when
the environment (probability/intensity of attack) is predictable (Clark and Harvell 1992, Åström and Lundberg 1994).
Our conclusion that latitudinal differences in palatability are constitutive is consistent with past studies
that have suggested that salt marsh plants may differentiate genetically across latitude, or in traits that
affect palatability to herbivores. O’Brien and Freshwater (1999) documented latitudinal variation in genetic structure of S. alterniflora, and Seliskar et al.
(2002) demonstrated that geographic differences in a
variety of traits of S. alterniflora persist over six years
in a common-garden environment. Finally, invasive
populations of S. alterniflora in Willapa Bay, California that have been without herbivores for many
generations appear to lack genetic resistance to herbivory, either because of founder effects or lack of
selection for resistance (Daehler and Strong 1997,
June 2005
LATITUDINAL VARIATION IN PALATABILITY
Garcia-Rossi et al. 2003). Thus, it is consistent with
these results to conclude that variation in palatability
of salt marsh plants across latitude might be under
strong genetic control.
Given the evidence that latitudinal differences in
plant palatability in coastal salt marshes are constitutive, it is of interest to determine what plant traits
may be the proximate determinants of this variation
in palatability. A variety of plant traits including
toughness, nitrogen and mineral content, and secondary metabolites may affect palatability to consumers
(Pennings and Paul 1992, Pennings et al. 1998), and
any of these might differ across latitude. Because it
is logistically difficult to measure every single trait
that might matter to herbivores, we centered our attention on toughness and nitrogen content, both of
which are relatively easy to measure. Leaf toughness
has been shown to be an efficient anti-herbivore defense (Feeny 1970, Coley 1983, 1987, Raupp 1985).
Nitrogen content is an important component of leaf
nutritional quality (White 1978, 1993), and several
studies report it to affect herbivore choices within a
plant species (Vince et al. 1981, Denno et al. 1986,
Bowdish and Stiling 1998, Gratton and Denno 2003),
with N-rich plants appearing to be more attractive to
herbivores than N-poor plants.
Using plant extracts and fresh and reconstituted plant
material, Siska et al. (2002) found that no single trait
explained latitudinal variation in palatability for all salt
marsh plants. In regards to the plants studied here,
toughness varied as a function of latitude for Spartina,
and nitrogen content varied as a function of latitude
for Distichlis and Spartina (Siska et al. 2002). In agreement with Siska et al. (2002), we found latitudinal variation in toughness and nitrogen content of Spartina,
no latitudinal variation in toughness of Solidago, and
inconclusive patterns in toughness of Distichlis. In contrast to Siska et al. (2002), we did not find latitudinal
variation in nitrogen content of Distichlis, although in
both studies there was considerable temporal variation
in the strength of the latitudinal differences. Finally,
we found clear latitudinal variation in nitrogen content
of Solidago, whereas Siska et al.’s data were inconclusive for this species, with their early-season comparisons showing stronger differences than late-season
comparisons, likely because late-season plants were
flowering and rosette leaves beginning to senesce. Although these results continue to shed light on latitudinal
differences in plant traits, we did not perform experiments in which we manipulated traits to address consumer responses, and so our understanding of the relationship between latitudinal differences in traits and
differences in palatability remains correlative rather
than causal. Importantly, however, when we found differences in plant traits (e.g., toughness and N in Spartina), these differences persisted across clonal generations in the common-garden environment, again sug-
1577
gesting strong genetic control of plant traits across latitude.
Our results come with three potential caveats. First,
feeding assays were conducted with consumers collected only from low-latitude sites. Pennings et al.
(2001), however, found that both high- and low-latitude
salt marsh herbivores prefer high- versus low-latitude
plants, and so it is unlikely that the origin of consumers
affected our results. Second, the plant propagation process and palatability assays were not replicated in
greenhouses at both high- and low-latitude sites. Our
greenhouse environment was not a match to either native environment, and was probably less physically
stressful than either (see Methods), but it is conceivable
that results might have differed in some other greenhouse environment (Haukioja et al. 1983). Results of
a pilot experiment with Spartina alterniflora; however,
suggest that this prospect is not likely. We grew 12
high-latitude (Rhode Island) and 12 low-latitude (Georgia) clones of S. alterniflora to the third clonal generation in a Rhode Island greenhouse. After two years,
high-latitude plants were strongly preferred by the
grasshopper Orchelimum over low-latitude clones (n 5
12, t 5 6.74, P , 0.0001), just as reported here for
plants grown in a Houston greenhouse. Third, because
we only assessed the palatability of one generation of
Solidago, the respective results might still reflect the
conditions experienced by the parental generation.
However, because maternally induced resistance to herbivores is short-lived in progeny plants, rapidly relaxing in the face of low herbivore pressure (Agrawal
2002), we think it unlikely that our results were influenced by maternal effects.
Given that latitudinal differences in palatability and
plant traits exist, we can speculate about the selective
forces that create and maintain this variation. Our work
has been presented in the context of standard biogeographic theory, arguing that consumer-prey interactions
are more intense at lower latitudes, and hence that prey
defenses should be better developed at lower latitudes.
Some evidence (Pennings et al. 2001; S. C. Pennings,
C. S. Salgado, and C. K. Ho, unpublished data) does
suggest that in Atlantic Coast salt marshes, herbivore
densities and damage to plants are greater at low vs.
high latitudes. This, however, would not exclude other
factors from selecting for geographic variation in plant
palatability. For example, differences in growing season length are correlated with leaf toughness (Wright
et al. 2004), and hence might affect plant palatability
to herbivores. Similarly, N content increases with latitude across a wide variety of plant taxa, likely as a
response to geographic variation in growing season
length and soil nutrient availability (Reich and Oleksyn
2004), and these differences in leaf N might affect plant
palatability to herbivores. Thus, although our results
are consistent with the theory that herbivore pressure
selects for latitudinal variation in plant palatability, we
cannot discard the possibility that other selective pres-
1578
CRISTIANO S. SALGADO AND STEVEN C. PENNINGS
sures are simultaneously mediating this latitudinal pattern of palatability.
Understanding the genetic and environmental bases
of intraspecific variation and how they covary on broad
geographic scales is an important goal of evolutionary
ecology, dating back to seminal work by Clausen et al.
(1940). Such information can provide important clues
to how organisms adapt to different and changing environments. Our results suggest that latitudinal variation in the palatability of Atlantic Coast salt marsh
plants, and the traits that might mediate palatability,
are constitutive, and thus likely under genetic control.
Although we did not explicitly test for induced defenses, our experiments demonstrated that plastic responses to herbivory, if present, are small compared
with the dramatic constitutive differences in plant palatability observed in the field across latitude.
ACKNOWLEDGMENTS
We thank B. Cole, E. Siemann, and G. Wellington for helpful suggestions, N. Davé, C.-K. Ho, L. Iwahara, J. Jiménez,
L. Johnson, A. Kunza, and C. McNutt for assistance in the
field and/or greenhouse, E. Siska for conducting pilot experiments with Spartina, and Bruce McNamee from Conrad Fafard, Inc. for providing potting soil. We thank NSF (DEB
0296160) and the Georgia Coastal Ecosystems LTER
(OCE99-82133) for funding. J. Cronin, E. Haukioja and an
anonymous reviewer provided helpful comments that improved the manuscript. This is contribution number 943 from
the University of Georgia Marine Institute.
LITERATURE CITED
Agrawal, A. A. 1998. Induced responses to herbivory and
increased plant performance. Science 279:1201–1202.
Agrawal, A. A. 2002. Herbivory and maternal effects: mechanisms and consequences of transgenerational induced
plant resistance. Ecology 83:3408–3415.
Agrawal, A., J. Conner, M. Johnson, and R. Wallsgrove. 2002.
Ecological genetics of an induced plant defense against
herbivores: additive genetic variance and costs of phenotypic plasticity. Evolution 56:2206–2213.
Åström, M., and P. Lundberg. 1994. Plant defense and stochastic risk of herbivory. Evolutionary Ecology 8:288–298.
Bakus, G. J., and G. Green. 1974. Toxicity in sponges and
holothurians: a geographic pattern. Science 185:951–953.
Bolser, R. C., and M. E. Hay. 1996. Are tropical plants better
defended? Palatability and defenses of temperate vs. tropical seaweeds. Ecology 77:2269–2286.
Bowdish, T. I., and P. Stiling. 1998. The influence of salt and
nitrogen on herbivore abundance: direct and indirect effects. Oecologia 113:400–405.
Callaway, R. M., S. C. Pennings, and C. L. Richards. 2003.
Phenotypic plasticity and interactions among plants. Ecology 84:1151–1128.
Clark, C. W., and C. D. Harvell. 1992. Inducible defenses
and the allocation of resources: a minimal model. American
Naturalist 139:521–539.
Clausen, J., D. D. Keck, and W. Hiesey. 1940. Experimental
studies on the nature of species. I. Effects of varied environments on western North American plants. Publication
520. Carnegie Institution of Washington, Washington,
D.C., USA.
Coley, P. D. 1983. Herbivory and defensive characteristics
of tree species in a lowland tropical forest. Ecological
Monographs 53:209–233.
Ecology, Vol. 86, No. 6
Coley, P. D. 1987. Patrones en las denfensas de las plantas:
Por que los herbivoros prefieren ciertas especies? Revista
de Biologia Tropical 35(Supplement 1):151–164.
Coley, P. D., and T. M. Aide. 1991. Comparison of herbivory
and plant defenses in temperate and tropical broad-leaved
forests. Pages 25–49 in P. W. Price, editor. Plant–animal
interactions: evolutionary ecology in tropical and temperate regions. John Wiley and Sons, New York, New
York, USA.
Coley, P. D., J. P. Bryant, and F. S. Chapin, III. 1985. Resource availability and plant antiherbivore defense. Science
230:895–899.
Daehler, C. C., and D. R. Strong. 1997. Reduced herbivore
resistance in an introduced smooth cordgrass (Spartina alterniflora) after a century of herbivore-free growth. Oecologia 110:99–108.
Denno, R. F., L. W. Douglass, and D. Jacobs. 1986. Effects
of crowding and host plant nutrition on a wing-dimorphic
planthopper. Ecology 67:116–123.
Feeny, P. 1970. Seasonal changes in oak leaf tannins and
nutrients as a cause of spring feeding by winter moth caterpillar. Ecology 51:565–581.
Feeny, P. 1976. Plant apparency and chemical defense. Pages
1–40 in J. Wallace and R. L. Mansell, editors. Biochemical
interaction between plants and insects. Recent advances in
phytochemistry. Volume 10. Plenum Press, New York, New
York, USA.
Garcia-Rossi, D., N. Rank, and D. R. Strong. 2003. Potential
for self-defeating biological control? Variation in herbivore
vulnerability among invasive Spartina genotypes. Ecological Applications 13:1640–1649.
Gratton, C., and R. F. Denno. 2003. Inter-year carryover effects of a nutrient pulse on Spartina plants, herbivores, and
natural enemies. Ecology 84:2692–2707.
Haukioja, E., K. Kapianinen, P. Niemelä, and J. Tuomi. 1983.
Plant availability hypothesis and other explanations of herbivore cycles: complementary or exclusive alternatives?
Oikos 40:419–432.
Hay, M. E., and W. Fenical. 1988. Marine plant–herbivore
interactions: the ecology of chemical defense. Annual Review of Ecology and Systematics 19:111–145.
Jablonski, D. 1993. The tropics as a source of evolutionary
novelty through geological time. Nature 364:142–144.
Jeanne, R. L. 1979. A latitudinal gradient in rates of ant
predation. Ecology 60:1211–1224.
Karban, R., and I. T. Baldwin. 1997. Induced responses to
herbivory. University of Chicago Press, Chicago, Illinois,
USA.
Levin, D. A. 1976. Alkaloid-bearing plants: an ecographical
perspective. American Naturalist 110:261–284.
MacArthur, R. H. 1972. Geographical ecology: patterns in
the distribution of species. Harper and Row, New York,
New York, USA.
O’Brien, D. L., and D. W. Freshwater. 1999. Genetic diversity
within tall form Spartina alterniflora Loisel. along the Atlantic and Gulf coasts of the United States. Wetlands 19:
352–358.
Pennings, S. C., and M. D. Bertness. 1999. Using latitudinal
variation to examine effects of climate on coastal salt marsh
pattern and process. Current Topics in Wetland Biogeochemistry 3:100–111.
Pennings, S. C., and M. D. Bertness. 2001. Salt marsh communities. Pages 289–316 in M. D. Bertness and M. E. Hay,
editors. Marine community ecology. Sinauer, Sunderland,
Massachusetts, USA.
Pennings, S. C., T. H. Carefoot, E. L. Siska, M. E. Chase,
and T. A. Page. 1998. Feeding preferences of a generalist
salt marsh crab: relative importance of multiple plant traits.
Ecology 79:1968–1979.
June 2005
LATITUDINAL VARIATION IN PALATABILITY
Pennings, S. C., and V. J. Paul. 1992. Effect of plant toughness, calcification, and chemistry on herbivory by Dolabella auricularia. Ecology 73:1606–1619.
Pennings, S. C., E. L. Siska, and M. D. Bertness. 2001. Latitudinal differences in plant palatability in Atlantic coast
salt marshes. Ecology 82:1344–1359.
Raupp, M. J. 1985. Effects of leaf toughness on mandibular
wear of the leaf beetle Plogiaodera versicolora. Ecological
Entomology 10:73–79.
Reich, P. B., and J. Oleksyn. 2004. Global patterns of plant
leaf N and P in relation to temperature and latitude. Proceedings of the National Academy of Sciences (USA) 101:
11001–11006.
Richards, C. L., J. L. Hamrick, L. A. Donovan, and R. Mauricio. 2004. Unexpectedly high clonal diversity of two salt
marsh perennials across a severe environmental gradient.
Ecology Letters 7:1155–1162.
Seliskar, D. M., J. L. Gallagher, D. M. Burdick, and L. A.
Mutz. 2002. The regulation of ecosystem functions by ecotypic variation in the dominant plant: a Spartina alterniflora
salt-marsh case study. Journal of Ecology 90:1–11.
Siska, E. L., S. C. Pennings, T. L. Buck, and M. D. Hanisak.
2002. Latitudinal variation in palatability of salt-marsh
plants: which traits are responsible? Ecology 83:3369–
3381.
Steinberg, P. D. 1992. Geographic variation in the interaction
between marine herbivores and brown algal secondary me-
1579
tabolites. Pages 51–92 in V. J. Paul, editor. Ecological roles
of marine natural products. Comstock Publishing Associates, Ithaca, New York, USA.
Steinberg, P. D., J. A. Estes, and F. C. Winter. 1995. Evolutionary consequences of food chain length in kelp forest
communities. Proceedings of the National Academy of Sciences (USA) 92:8145–8148.
Swihart, R. K., J. P. Bryant, and L. Newton. 1994. Latitudinal
patterns in consumption of woody plants by snowshoe hares
in the eastern United States. Oikos 70:427–434.
Van Alstyne, K. L., and V. J. Paul. 1990. The biogeography
of polyphenolic compounds in marine macroalgae: temperate brown algal defenses deter feeding by tropical herbivorous fishes. Oecologia 84:158–163.
Vermeij, G. J. 1978. Biogeography and adaptation. Harvard
University Press, Cambridge, Massachusetts, USA.
Vince, S. W., I. Valiela, and J. M. Teal. 1981. An experimental
study of the structure of herbivorous insect communities
in a salt marsh. Ecology 62:1662–1678.
White, T. C. R. 1978. The importance of a relative shortage
of food in animal ecology. Oecologia 33:71–86.
White, T. C. R. 1993. The inadequate environment: nitrogen
and the abundance of animals. Springer-Verlag, Berlin,
Germany.
Wright, I. J., et al. 2004. The worldwide leaf economics
spectrum. Nature 428:821–827.
APPENDIX A
A table listing collection sites and geographic coordinates is available in ESA’s Electronic Data Archive: Ecological Archives
E086-085-A1.
APPENDIX B
A color version of Plate 1 is available in ESA’s Electronic Data Archive: Ecological Archives E086-085-A2.