Content uploaded by Suzanne E. Tank
Author content
All content in this area was uploaded by Suzanne E. Tank
Content may be subject to copyright.
51
J. N. Am. Benthol. Soc., 2003, 22(1):51–62
q
2003 by The North American Benthological Society
Elemental composition of littoral invertebrates from oligotrophic and
eutrophic Canadian lakes
P
AUL
C. F
ROST
1
Department of Biology, Arizona State University, Tempe, Arizona 85287-1501 USA
S
UZANNE
E. T
ANK
Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9
M
ICHAEL
A. T
URNER
Experimental Lakes Area, Fisheries and Oceans Canada, 501 University Crescent, Winnipeg, Manitoba,
Canada R3T 2N6
J
AMES
J. E
LSER
Department of Biology, Arizona State University, Tempe, Arizona 85287-1501 USA
Abstract. The P, N, and C content of littoral macroinvertebrates from 8 lakes located in 3 distinct
geographical regions of Canada (central Alberta, northwestern Ontario, and Rocky Mountains near
Jasper, Alberta) is described. A wide range of values was found in the body content (all values are
% of dry mass) of P (0.4–1.6%), N (5.8–13.7%), and C (32.5–53.5%) in the data set containing inver-
tebrates from all 8 lakes. C:P (63–324), N:P (9.6–60), and C:N (4.2–7.6) (all by atom) also varied widely.
This variation was partly related to the different mean body content of P, N, and C among taxonomic
groups. However, the mean P, N, and C content of macroinvertebrate assemblages varied little among
lakes. The patterns of elemental composition in benthic invertebrates shown here are similar to zoo-
plankton and terrestrial insects, and indicate that the strength of stoichiometric constraints acting in
littoral food webs will depend on the taxa being considered.
Key words: C:N:P ratios, benthos, body stoichiometry, aquatic insects.
The study of ecological stoichiometry eluci-
dates how organisms affect and are affected by
the balance of energy and multiple chemical el-
ements in ecosystems (Reiners 1986, Frost et al.
2002, Sterner and Elser 2002). One focus of eco-
logical stoichiometry is the grazer–producer in-
terface, where C:nutrient ratios in food often
dramatically exceed those found in consumers
(Sterner and Hessen 1994, Elser et al. 2000a).
These elemental imbalances between herbivores
and their food can strongly affect organism nu-
trient release, growth, and reproduction with
subsequent consequences for population dy-
namics (Urabe et al. 2002) and ecosystem pro-
cesses (Hessen 1997, Sterner et al. 1997). How-
ever, to estimate the extent that elemental con-
straints act on consumers minimally requires
knowledge of the C:nutrient ratios in their food
1
Present address: Department of Biological Scienc-
es, University of Notre Dame, Notre Dame, Indiana
46556 USA. E-mail: periphyton@yahoo.com
sources and those needed for their growth, re-
production, and maintenance (Sterner and
Schulz 1998, Frost and Elser 2002). The body
content of P, N, and C in consumers is one in-
dicator of their demands for these elements. Be-
cause the C:N:P ratios of many littoral macro-
invertebrates remain unknown, there is yet little
basis to apply recent developments in ecological
stoichiometry to studies of ecological processes
in littoral systems (Steinman 1996, Frost et al.
2002).
Some insight into patterns of P, N, and C con-
tent of freshwater invertebrates can be gained
from a relatively well-described group, meta-
zoan zooplankton (Elser et al. 2000a). In zoo-
plankton, little variability in C:N:P ratios was
originally reported within taxonomic groups,
and large differences were found among taxa
(Andersen and Hessen 1991, Hessen and Lyche
1991). Interspecific differences in zooplankton
elemental composition can be explained at com-
plementary levels that link organismal physiol-
52 [Volume 22P. C. F
ROST ET AL
.
T
ABLE
1. Characteristics of lakes sampled in our benthic invertebrate surveys.
Lake Date sampled Latitude Longitude
Total P
(
m
g/L)
Coal
Hastings
Nakamun
Thunder
239
373
Hibernia
Honeymoon
26 June 2000
19 June 2000
23 June 2000
21 June 2000
04 July 2000
05 July 2000
19 July 2000
20 July 2000
53
8
08
9
N
53
8
25
9
N
53
8
53
9
N
54
8
09
9
N
49
8
40
9
N
49
8
44
9
N
52
8
55
9
N
52
8
55
9
N
113
8
21
9
W
113
8
55
9
W
114
8
12
9
W
114
8
45
9
W
93
8
43
9
W
93
8
48
9
W
118
8
05
9
W
118
8
05
9
W
176
a
136
b
88
c
45
d
5.8
e
4.1
f
11
g
6.5
g
a
Mean concentration for the summer of 1984 (Mitchell and Prepas 1990)
b
Mean concentration for the summers of 1981 and 1982 (Mitchell and Prepas 1990)
c
Mean concentration for the summers of 1982 and 1983 (Mitchell and Prepas 1990)
d
Mean concentration for the summers of 1984, 1986, and 1988 (Mitchell and Prepas 1990)
e
Mean concentration for the summers 1981 to 1998 (M. Stainton, Freshwater Institute, Winnipeg, Manitoba,
unpublished data)
f
Mean concentration for the summers of 1994 to 1997 (M. Stainton, unpublished data)
g
Mean concentration for the summer of 2000 (SET, unpublished data)
ogy and evolution (Elser et al. 1996, Dobberfuhl
1999). For example, interspecific variation in
zooplankton body C:N:P ratios reflect: 1) the
relative contribution of primary cellular biomol-
ecules, and 2) different life-history strategies
that result from costs associated with maintain-
ing P-rich body stoichiometry (Elser et al. 1996,
2000b). However, before similar integrative hy-
pothesis testing can take place for lake benthic
consumers, descriptive studies need to docu-
ment the patterns of C:N:P ratios within and
among taxa from widely varying environments.
We examine patterns of elemental composi-
tion in littoral consumer assemblages from lakes
across a trophic gradient. Based on recent stud-
ies of zooplankton and terrestrial insects (e.g.,
Andersen and Hessen 1991, Fagan et al., in
press), we expected to find significant differenc-
es among taxonomic groups. We were also in-
terested in whether these patterns would be
consistent in littoral invertebrates of different
trophic levels collected from eutrophic and oli-
gotrophic lakes. We use these data to forecast
which aquatic macroinvertebrates would most
likely experience strong elemental imbalances
with respect to their food.
Methods
Study sites
Benthic invertebrates were sampled from
lakes located in 3 distinct geographical areas of
Canada, encompassing a wide range of physi-
cochemical conditions (Table 1). Nakamun, Has-
tings, Coal, and Thunder lakes lie on the Boreal
Plain near Edmonton, Alberta, and were chosen
to span meso- to hypereutrophic conditions (to-
tal P from 45–175
m
g P/L, Table 1). Lakes in this
area generally have high nutrient concentrations
and littoral zones dominated by macrophytes
growing on soft sediments (Mitchell and Prepas
1990). Lakes (L) 239 and 373, sampled at the
Experimental Lakes Area in northwestern On-
tario, are oligotrophic with littoral zones com-
posed primarily of hard granite (Schindler et al.
1973). Hibernia and Honeymoon Lakes in Jasper
National Park in the Canadian Rockies of west-
ern Alberta were also sampled. Littoral zones of
these oligotrophic lakes have a mixture of sand
and rock substrates.
Collection methods
Littoral zone sampling was aimed at collect-
ing a wide array of invertebrate types from dif-
ferent trophic levels. Each lake was intensively
sampled once between mid June and late July in
the summer of 2000. Samples were repeatedly
taken from 1 or 2 haphazardly chosen areas in
each lake with a hand-held net. Additional
leeches and caddisflies were collected by hand.
Samples were collected until adequate numbers
of each taxonomic group were acquired for ele-
mental analysis. Invertebrates were sorted lake-
2003] 53E
LEMENTAL COMPOSITION OF LITTORAL MACROINVERTEBRATES
T
ABLE
2. Invertebrate taxa collected from central Alberta lakes. The family and, if possible, genus (in pa-
rentheses) of each specimen was determined. Amphipods were identified to species, whereas leeches were not
identified past order. Specimens were not found (n.f.) for some taxonomic groups in some lakes.
Order Coal Lake Hastings Lake Nakamun Lake Thunder Lake
Amphipoda Hyalella azteca Hyalella azteca
1
Gammarus lacustris
Hyalella azteca Hyalella azteca
1
Gammarus lacustris
Coleoptera Dytiscidae
(Laccophilus)
Hydrophilidae Dytiscidae
(Desmopachria)
Dytiscidae
Diptera
Ephemeroptera
Chironomidae
Caenidae
(Caenis)
Chironomidae
Caenidae
(Caenis)
Chironomidae
Caenidae
(Caenis)
Chironomidae
Caenidae
(Caenis)
Hemiptera
Hirudinea
Odonata, Anisoptera
Odonata, Zygoptera
Trichoptera
Corixidae
–
n.f.
Coenagrionidae
(Enallagma)
Leptoceridae
(Oecetis)
Corixidae
–
n.f.
Coenagrionidae
(Enallagma)
Leptoceridae
(Oecetis)
Corixidae
–
n.f.
Coenagrionidae
(Enallagma)
Leptoceridae
(Oecetis)
Corixidae
–
n.f.
Coenagrionidae
(Enallagma)
Leptoceridae
(Oecetis)
T
ABLE
3. Invertebrate taxa collected from lakes at the Experimental Lakes Area and in Jasper National Park.
The family and, if possible, genus (given in parentheses) of each specimen was determined. Amphipods were
identified to species, whereas leeches were not identified past order. Specimens were not found (n.f.) for some
taxonomic groups in some lakes.
Order L373 L239 Honeymoon Lake Hibernia Lake
Amphipoda Hyalella azteca Hyalella azteca Hyalella azteca
1
Gammarus lacustris
Hyalella azteca
Coleoptera n.f. Dytiscidae
(Hydroporus)
n.f. n.f.
Diptera
Ephemeroptera
Chironomidae
Baetidae
(Callibaetis)
Chironomidae
Siphlonuridae
(Siphlonurus)
Chironomidae
Caenidae
(Caenis)
Chironomidae
Caenidae
(Caenis)
Hemiptera
Hirudinea
Odonata, Anisoptera
n.f.
–
Gomphidae
(Gomphus)
Corixidae
–
Gomphidae
n.f.
–
Libellulidae
(Pachydiplax)
Corixidae
–
Aeschnidae
(Aeshna)
Odonata, Zygoptera n.f. n.f. n.f. Coenagrionidae
(Enallagma)
Trichoptera Limnephilidae Limnephilidae Limnephilidae
(Limnephilus)
Limnephilidae
(Limnephilus)
side, placed into polyethylene bags containing
lake water, and stored on ice in the dark. Ani-
mals were transported back to the laboratory
and held for at least 24 h to allow gut clearance.
Invertebrates were placed onto aluminum pans,
dried for
.
48 h, and stored in a freezer until
they were processed for elemental analysis. Ad-
ditional invertebrates were preserved in 80%
ethanol until they were identified under a dis-
secting microscope to family or, when possible,
genus and species. Our survey included a wide
array of benthic taxa from 8 different orders (Ta-
bles 2, 3).
Chemical analyses
After drying, individual large animals (
.
3
mg) were ground into a fine powder from
which subsamples (
;
1 mg) were weighed. To
obtain enough material from smaller animals
(
,
3 mg), individuals were pooled as needed to
yield
.
3 mg total mass. Grouped samples were
54 [Volume 22P. C. F
ROST ET AL
.
F
IG
. 1. Frequency distribution of body P, N, and C content (% of dry mass) and their molar elemental ratios
(C:N, C:P, and N:P) from benthic invertebrate assemblages sampled in this study.
also ground to a fine powder before subsamples
(
;
1 mg) were weighed for elemental analysis.
For C and N analysis, weighed subsamples were
placed into tin capsules and stored under des-
iccation. Samples were analyzed for C and N
content on a Finnigan Delta
1
mass spectrome-
ter at the University of Arkansas Stable Isotope
Laboratory (Fayetteville, Arkansas). Phosphorus
was estimated from subsamples of the remain-
ing material after digestion with potassium per-
sulfate by autoclaving for
.
30 min using the
molybdate-blue reaction (APHA 1992).
Statistical and data analysis
For each lake, differences in mean P, N, and
C content among sampled taxa were tested us-
ing 1-way ANOVA (SAS Institute Inc. 1987.
SAS/STAT user’s guide, 6
th
edition, SAS Insti-
tute Inc., Cary, North Carolina), followed by a
posteriori comparisons (Tukey’s HSD test).
Weighted ranks of each taxon were also calcu-
lated to compare within-lake patterns of inver-
tebrate P, N, and C content among lakes.
Weighted ranks were calculated in each lake by
ranking invertebrates from lowest to highest by
their nutrient content. This rank was divided by
the number of ranked groups in that lake and
multiplied by the average number of taxa in all
lakes. Taxonomic differences among average
weighted ranks were tested using 1-way ANO-
VA (SAS Institute Inc. 1987), followed by a pos-
teriori comparisons among means (Tukey’s HSD
test).
Results
The P, N, and C content of benthic macroin-
vertebrates varied considerably in animals col-
lected from the littoral zones of our 8 study
lakes (Fig. 1). Body P content from all taxa sam-
pled ranged
.
3-fold from 0.41% to 1.40% (Fig.
1). Body N varied 2-fold from 5.77% to 13.6%.
Body C content was less variable, ranging only
from 32.5% to 53.5%. Body C content had a co-
efficient of variation (CV) of 12.2% compared to
20.8% for P and 18.3% for N (Fig. 1). C:P, N:P,
and C:N molar ratios also varied widely (Fig. 1).
The mean ratios for benthic invertebrates found
2003] 55E
LEMENTAL COMPOSITION OF LITTORAL MACROINVERTEBRATES
T
ABLE
4. Comparison of descriptive statistics for
C:P, N:P, and C:N ratios from benthic invertebrates
sampled in this study with those from freshwater zoo-
plankton and terrestrial invertebrates. Freshwater zoo-
plankton and terrestrial invertebrate data obtained
from Elser et al. (2000a).
Ratio
Benthic
inverte-
brates
Fresh-
water
zoo-
plankton
Terres-
trial
inverte-
brates
C:P n
Mean
Median
SD
CV (%)
68
148
141
51.2
34
40
124
114
48
38
27
116
73.2
72.4
62
N:P n
Mean
Median
SD
CV (%)
68
27.3
25.8
9.92
36
37
22.3
18.5
10.5
47
22
26.4
22.6
10.1
38
C:N n
Mean
Median
SD
CV (%)
68
5.6
5.5
0.75
13
38
6.3
6.0
1.3
21
124
6.5
6.4
1.9
29
here were similar to those shown previously for
zooplankton and terrestrial insects (Table 4).
Variation in invertebrate body P content re-
sulted partly from significant differences among
various littoral taxa. Significant differences were
found in the P content of at least some taxonom-
ic groups in all but one lake (Nakamun) (Fig. 2).
In addition, taxonomic groups showed similar
patterns of in-lake taxonomic rankings of P con-
tent. For example, beetles and leeches were rel-
atively P-poor because both had low weighted
ranks (Fig. 3). Small amphipods were relatively
P-rich, with a high weighted rank (Fig. 3). May-
flies, caddisflies, hemipterans, damselflies,
dragonflies, and midges all had intermediate P
content and variability in their ranking among
lakes (Fig. 3). The mean P content of inverte-
brates across all lakes reflected these in-lake
rankings. Amphipods, hemipterans, and may-
flies generally had the highest mean P content,
whereas leeches and beetles had lowest mean P
content (Table 5).
Body N content also showed significant dif-
ferences among taxa (Fig. 4). In-lake ranking of
invertebrate N content was generally opposite
to those based on P content. Leeches, hemipter-
ans, dragonflies, and damselflies were relatively
N-rich as indicated by their high average ranks
(Fig. 3). N-poor taxa included large and small
amphipods, caddisflies, mayflies, and midges,
which all showed low in-lake rankings (Fig. 3).
Consequently, there were differences in the N
content among invertebrates when averaged
across lakes. Amphipods, caddisflies, mayflies,
and midges had relatively low N content,
whereas leeches, hemipterans, and damselflies
were the most N-rich (Table 5).
Body C content was generally similar among
taxa with the exception of the amphipods (Fig.
5, Table 5). Amphipods, both large and small,
had significantly lower body C content than
other taxa in all lakes (Figs 3, 5). Otherwise, few
differences were found among non-amphipod
benthic invertebrates in body C content.
C:P, N:P, and C:N ratios reflected taxonomic
differences in P, N, and C content of various lit-
toral invertebrates. For example, C:P ratios were
much lower in amphipods than in beetles when
averaged across lakes (Table 5). In addition, rel-
atively greater differences were found among
taxa for C:P and N:P ratios compared to C:N
ratios. When comparing mean ratios among dif-
ferent taxonomic groups, C:P and N:P ratios
ranged 2.7 and 2.1 times, respectively, com-
pared to 1.4 for C:N ratios (Table 5).
Little variation was found among the within-
lake averages of P, N, and C content of benthic
invertebrates (Table 6). For example, inverte-
brates from a P-poor lake, Honeymoon, had the
same mean P content as those in a P-rich lake,
Coal. Similarly, mean C:P, N:P, and C:N ratios
of benthic invertebrates were relatively invariant
among lakes (Table 6).
Discussion
The first detailed examination of the elemen-
tal composition of benthic macroinvertebrates
from lakes differing in nutrient supply is pre-
sented here. Despite a large trophic gradient,
where P ranged
.
40-fold (Table 1), few differ-
ences in within-lake averages of invertebrate P,
N, or C content were found among lakes (Table
6). In addition, benthic macroinvertebrates col-
lected from these 8 lakes were similar in their
average elemental composition compared to
other lake and terrestrial consumers (Table 4).
Consequently, the average requirements of ele-
56 [Volume 22P. C. F
ROST ET AL
.
F
IG
. 2. Percent body P content (mean
1
SD) in different taxonomic groups of benthic invertebrates according
to lake. For each lake, differences among taxonomic groups were analyzed using 1-way ANOVA. Significant
ANOVAs ( p
,
0.05) were followed with multiple comparisons (Tukey’s HSD) among means. Taxa with different
letters were significantly different from each other. Animal taxa are abbreviated as follows: Amphipoda (A, l.
and s.
5
large and small, respectively), Anisoptera (Ani, l. and s.
5
large and small, respectively), Coleoptera
(Co), Diptera (Dip), Ephemeroptera (Eph), Hemiptera (Hem), Hirudinea (Hir), Trichoptera (Tri), and Zygoptera
(Zyg). L
5
Lake.
ments needed for growth and reproduction ap-
pear to be relatively similar among metazoan
consumers from widely diverse environments.
As such, mass balance constraints that act on
consumer physiology and population dynamics
in pelagic (Sterner and Elser 2002) and terres-
trial (Higashi et al. 1992) ecosystems are also
likely to apply to the benthic zones of lakes.
2003] 57E
LEMENTAL COMPOSITION OF LITTORAL MACROINVERTEBRATES
F
IG
. 3. Mean (across all lakes) weighted ranking of benthic invertebrates by their P, N, and C contents. See
Methods for a description of the derivation of the weighted ranks. A low average rank indicates that a taxon has
a relatively low content of that particular element compared to other taxa. A large error bar (SD) indicates the
taxon was highly ranked in some lakes but not others. Differences among taxonomic groups were analyzed using
1-way ANOVA. Significant ANOVAs ( p
,
0.05) were followed with multiple comparisons (Tukey’s HSD) among
means. Taxa with different letters were significantly different from each other. Abbreviations as in Fig. 2.
Proximate and ultimate causes of organismal
elemental composition
Significant differences in C, N, and P content
among different littoral consumer taxa were ap-
parent. Such taxonomic differences in C, N, and
P content have been linked to differences in the
body content of important pools of biomole-
cules (Elser et al. 1996, 2000c). C:N:P ratios in
benthic macroinvertebrates likely reflect, in part,
the relative contribution of structural molecules
(C- and N-rich) and RNA (P-rich) to body tis-
sues (Elser et al. 1996). For example, the rela-
tively high N content of leeches may result from
high structural protein content, given the lack
of other potentially important biomolecules
(e.g., chitin, Brown 1975) in their bodies. On the
58 [Volume 22P. C. F
ROST ET AL
.
T
ABLE
5. Elemental composition (mean and SD) of different benthic invertebrates averaged across the sam-
pled lakes. n
5
number of lakes i which each taxon was found.
Amphi-
poda
Trichop-
tera
Ephemer-
optera Diptera Hirudinea Zygoptera
Hemip-
tera
Coleop-
tera
Anisop-
tera
n88888 5666
P Mean
SD
1.01
0.11
0.86
0.24
0.92
0.11
0.83
0.12
0.77
0.10
0.78
0.13
1.00
0.32
0.54
0.16
0.96
0.12
N Mean
SD
7.28
0.68
8.48
0.66
9.83
0.60
9.52
0.85
12.1
1.03
10.9
1.22
11.7
0.85
10.8
0.17
11.0
0.39
C Mean
SD
36.2
2.22
49.7
2.11
47.3
1.98
47.7
4.34
49.5
1.44
47.5
2.33
50.7
4.24
50.4
0.70
48.3
1.45
C:P Mean
SD
93.8
12.3
157.6
40.8
133.2
14.1
151.0
29.9
168.8
18.4
162.2
31.7
142.3
47.5
253.9
63.7
130.9
15.9
N:P Mean
SD
16.2
2.45
24.2
4.43
23.7
2.92
25.9
5.53
35.4
3.99
31.8
4.90
27.9
8.32
46.7
11.7
25.7
3.54
C:N Mean
SD
5.84
0.61
6.90
0.53
5.62
0.24
5.86
0.51
4.79
0.42
5.13
0.94
5.07
0.31
5.44
0.14
5.11
0.12
other hand, relatively P-rich amphipods proba-
bly have less structural material and greater
RNA content.
Descriptive studies are needed to verify these
proposed relationships between C:N:P ratios
and biomolecular content in benthic inverte-
brates. First, biochemical composition needs to
be compared among benthic invertebrates that
have contrasting C:N:P ratios. Second, experi-
mental studies are also needed to examine the
extent that C:N:P ratios vary within macroin-
vertebrate taxa as a function of body size or life
stage. For example, Frost and Elser (2002) found
a negative relationship between P content and
body size in a littoral mayfly. Ontogenic chang-
es in body elemental composition and their
physiological causes remain largely unstudied
in benthic macroinvertebtrates. From a different
perspective, similarities among various taxo-
nomic groups may partly reflect shared ances-
try and/or functional characteristics, such as
feeding mode (Fagan et al., in press). However,
the extent that these factors may explain pat-
terns of invertebrate P, N, and C content, as
shown in our study, is unknown.
Complementary explanations of inter-taxo-
nomic differences in animal C:N:P composition
are based on the important roles that certain
biomolecules play in growth and reproduction.
Thus, organismal elemental composition reflects
selective pressures and the elemental require-
ments of fast growth and reproduction. Selec-
tion on alternative life-history strategies, which
require different allocations of cellular materials
(Elser et al. 1996, 2000b), may result in distinct
biomolecular composition (and hence, elemental
composition) of various taxa. For example, a
rapid-growth strategy appears inseparable from
a body rich in P arising from the high P content
of RNA, which is required for DNA replication
and protein expression (Elser et al. 1996, 2000c).
Conversely, a slower-growth lifestyle may in-
volve greater allocation to protective tissues,
which if rich in C and/or N would yield higher
C:P and N:P ratios in animal bodies. However,
similar connections between body C:N:P ratios
and life-history parameters remain to be deter-
mined in benthic macroinvertebrates.
Differential invertebrate susceptibility to poor food
quality
Differences in elemental composition among
taxa indicate that some benthic consumers will
more be susceptible than others to the effects of
nutrient-poor food (Frost et al. 2002). Based on
body P content alone, amphipods appear most
probable to experience P constraints on their
growth and reproduction. Conversely, leeches
and beetles would be least probable to experi-
ence P constraints given their relatively low
body P content. Actual susceptibility to P-limi-
tation will also be influenced by other factors
such as feeding rate and body physiology that
2003] 59E
LEMENTAL COMPOSITION OF LITTORAL MACROINVERTEBRATES
F
IG
. 4. Percent body N content (mean
1
SD) in different taxonomic groups of benthic invertebrates accord-
ing to lake. For each lake, differences among taxonomic groups were analyzed using 1-way ANOVA. Significant
ANOVAs ( p
,
0.05) were followed with multiple comparisons (Tukey’s HSD) among means. Taxa with different
letters were significantly different from each other. Abbreviations as in Fig. 2.
affect the supply of various elements relative to
consumer demand (Frost and Elser 2002). For
example, increasing dominance of respiration
relative to somatic growth lowers the probabil-
ity of P-limited growth regardless of an animal’s
C:N:P content (Urabe and Watanabe 1992, Frost
and Elser 2002). Susceptibility to the effects of
poor food quality needs to be examined for ben-
thic taxa that differ in C:N:P content.
Effects of trophic position on elemental imbalances
between consumer and food
The trophic position of a consumer also af-
fects the probability that it will encounter stoi-
60 [Volume 22P. C. F
ROST ET AL
.
F
IG
. 5. Percent body C content (mean
1
SD) in different taxonomic groups of benthic invertebrates according
to lake. For each lake, differences among taxonomic groups were analyzed using 1-way ANOVA. Significant
ANOVAs ( p
,
0.05) were followed with multiple comparisons (Tukey’s HSD) among means. Taxa with different
letters were significantly different from each other. Abbreviations as in Fig. 2.
chiometric constraints on its growth and repro-
duction (Sterner and Hessen 1994). Predatory
and parasitic taxa, such as beetles and leeches,
would be least likely to encounter stoichiometric
constraints because of the similarity between
the elemental composition in their body tissues
and their food sources. On the other hand, de-
tritivores and herbivores, such as amphipods
and mayflies, would be more likely to encounter
stoichiometric constraints on their growth and
reproduction because their bodies generally
contain more P and N compared to terrestrial,
plant-derived material (Gosz et al. 1973, Elser et
al. 2000a) and biofilms (Kahlert 1998, Frost and
Elser 2002). As a result, rates and ratios of nu-
trient release will also likely vary among con-
sumers at different trophic levels. Future work
should assess the extent that elemental con-
straints act on different trophic levels. Also, we
should determine how these constraints (or lack
2003] 61E
LEMENTAL COMPOSITION OF LITTORAL MACROINVERTEBRATES
T
ABLE
6. Elemental composition (mean and SD) of benthic invertebrates from our study lakes. Parenthetical
values indicate mean total P (
m
g/L) in the epilimnion of each lake. Lakes are abbreviated as follows: L (Lake),
Honeymoon (Hon), Hibernia (Hib), Thunder (Thun), Nakamun (Naka), and Hastings (Hast). n
5
number of
taxonomic groups found in each lake.
L373
(3)
L239
(5)
Hon
(6)
Hib
(11)
Thun
(30)
Naka
(35)
Hast
(80)
Coal
(300)
n79889999
P Mean
SD
0.78
0.16
0.72
0.18
0.88
0.20
0.89
0.17
0.85
0.14
0.91
0.13
0.91
0.19
0.88
0.23
N Mean
SD
9.61
1.46
9.92
1.83
9.34
2.23
9.84
1.68
9.89
2.17
10.2
1.56
9.56
1.85
9.98
2.12
C Mean
SD
44.4
4.89
46.6
5.93
44.5
6.91
47.5
4.98
46.4
5.88
46.9
5.67
47.3
5.04
45.3
5.42
C:P Mean
SD
153
40.8
184
65.7
138
41.0
146
47.2
146
39.9
138
35.9
144
56.0
149
74.6
N:P Mean
SD
28.3
7.61
33.1
13.2
24.9
8.54
25.8
8.52
26.9
8.83
25.8
6.97
25.1
10.9
28.3
14.8
C:N Mean
SD
5.45
0.70
5.67
0.69
5.72
0.98
5.74
0.82
5.61
0.90
5.39
0.40
5.88
0.84
5.42
0.69
thereof) affect consumer-driven nutrient recy-
cling in benthic ecosystems.
Our results describe inter-taxonomic differ-
ences in the C:N:P ratios of benthic macroinver-
tebrates of lake littoral zones. As noted above,
these results point to many questions that re-
main unanswered. In summary, 2 broad groups
of questions need to be addressed in future ben-
thic studies: 1) what proximate and ultimate
causes lead to differences in the C:N:P content
of benthic invertebrate taxa, and 2) what are the
ecological consequences of differences in C:N:P
ratios between benthic consumers? Our study
has thus documented many patterns that point
to potentially important gaps in our knowledge
of the biology of benthic consumers and their
relationships with their environment.
Acknowledgements
We thank J. Cox from the Stable Isotope Lab-
oratory at the University of Arkansas for her
careful elemental analysis of our samples. In ad-
dition, we thank P. Weidman and M. Xenopou-
los for their assistance with field sampling and
M. Bowman for helpful comments that im-
proved the manuscript. Funds for this research
were partially provided by the NSF (DEB-
9725867) to J. J. Elser.
Literature Cited
A
NDERSEN
,T.,
AND
D. O. H
ESSEN
. 1991. Carbon, nitro-
gen, and phosphorus content of freshwater zoo-
plankton. Limnology and Oceanography 36:807–
814.
APHA (American Public Health Association). 1992.
Standard methods for the examination of water
and wastewater. 18
th
edition. American Public
Health Association, Washington, DC.
B
ROWN
, C. H. 1975. Structural materials in animals.
Wiley, New York.
D
OBBERFUHL
, D. R. 1999. Elemental stoichiometry in
crustacean zooplankton: phylogenetic patterns,
physiological mechanisms, and ecological conse-
quences. PhD Dissertation, Department of Biolo-
gy, Arizona State University, Tempe, Arizona.
E
LSER
, J. J., D. R. D
OBBERFUHL
,N.A.M
AC
K
AY
,
AND
J.
H. S
CHAMPEL
. 1996. Organism size, life history,
and N:P stoichiometry: towards a unified view of
cellular and ecosystem processes. BioScience 46:
674–684.
E
LSER
, J. J. , W. F. F
AGAN
,R.F.D
ENNO
,D.R.D
OBBER
-
FUHL
,A.F
OLARIN
,A.H
UBERTY
,S.I
NTERLANDI
,S.
S. K
ILHAM
,E.M
C
C
AULEY
,K.L.S
CHULZ
,E.H.
S
IEMANN
,
AND
R. W. S
TERNER
. 2000a. Nutritional
constraints in terrestrial and freshwater food
webs. Nature 408:578–580.
E
LSER
, J. J., W. J. O’B
RIEN
,D.R.D
OBBERFUHL
,
AND
T.
E. D
OWLING
. 2000b. The evolution of ecosystem
processes: growth rate and elemental stoichiom-
etry of a key herbivore in temperate and arctic
62 [Volume 22P. C. F
ROST ET AL
.
habitats. Journal of Evolutionary Biology 13:845–
853.
E
LSER
, J. J., R. W. S
TERNER
,E.G
OROKHOVA
,W.F.F
A
-
GAN
,T.A.M
ARKOW
,J.B.C
OTNER
,J.F.H
ARRISON
,
S. E. H
OBBIE
,G.M.O
DELL
,
AND
L. J. W
EIDER
.
2000c. Biological stoichiometry from genes to eco-
systems. Ecology Letters 3:540–550.
F
AGAN
, W. F., E. S
IEMANN
,C.M
ITTER
,R.F.D
ENNO
,A.
F. H
UBERTY
,H.A.W
OODS
,
AND
J. J. E
LSER
. 2002.
Nitrogen in insects: implications for trophic com-
plexity and species diversification. American Nat-
uralist (in press).
F
ROST
,P.C.,
AND
J. J. E
LSER
. 2002. Growth responses
of littoral mayflies to the phosphorus content of
their food. Ecology Letters 5:232–240.
F
ROST
, P. C., R. S. S
TELZER
,G.A.L
AMBERTI
,
AND
J. J.
E
LSER
. 2002. Ecological stoichiometry of trophic
interactions in the benthos: understanding the
role of C:N:P ratios in littoral and lotic habitats.
Journal of the North American Benthological So-
ciety 21:515–528.
G
OSZ
, J. R., G. E. L
IKENS
,
AND
F. H. B
ORMANN
1973.
Nutrient release from decomposing leaf and
branch litter in the Hubbard Brook forest, New
Hampshire. Ecological Monographs 43:173–191.
H
ESSEN
, D. O. 1997. Stoichiometry in food webs—Lot-
ka revisited. Oikos 79:195–200.
H
ESSEN
,D.O.,
AND
A. L
YCHE
. 1991. Inter- and intra-
specific variations in zooplankton element com-
position. Archiv fu¨ r Hydrobiologie 121:343–353.
H
IGASHI
,M.,T.A
BE
,
AND
T. P. B
URNS
. 1992. Carbon-
nitrogen balance and termite ecology. Proceed-
ings of the Royal Society of London B, Biological
Sciences 249:303–308.
K
AHLERT
, M. 1998. C:N:P ratios of freshwater benthic
algae. Archiv fu¨ r Hydrobiologie, Special Issues:
Advances in Limnolog y 51:105–114.
M
ITCHELL
,P.,
AND
E. P
REPAS
. 1990. Atlas of Alberta
Lakes. The University of Alberta Press, Edmon-
ton, Alberta.
R
EINERS
, W. A. 1986. Complementary models for eco-
systems. American Naturalist 127:59–73.
S
CHINDLER
,D.W.,V.E.F
ROST
,
AND
R. V. S
CHMIDT
.
1973. Production of epilithiphyton in two lakes of
the Experimental Lakes Area, northwestern On-
tario. Journal of the Fisheries Research Board of
Canada 30:1511–1524.
S
TEINMAN
, A. D. 1996. Effects of grazers on freshwater
benthic algae. Pages 341–373 in R. J. Stevenson,
M. L. Bothwell, and R. L. Lowe (editors). Algal
ecology: freshwater benthic ecosystems. Academ-
ic Press, San Diego.
S
TERNER
,R.W.,
AND
J. J. E
LSER
. 2002. Ecological stoi-
chiometry: the biology of elements from mole-
cules to the biosphere. Princeton University Press,
Princeton, New Jersey.
S
TERNER
, R. W., J. J. E
LSER
,E.J.F
EE
,S.J.G
UILDFORD
,
AND
T. H. C
HRZANOWSKI
. 1997. The light:nutrient
ratio in lakes: the balance of energy and materials
affects ecosystem structure and process. Ameri-
can Naturalist 150:663–684.
S
TERNER
,R.W.,
AND
D. O. H
ESSEN
. 1994. Algal nutri-
ent limitation and the nutrition of aquatic herbi-
vores. Annual Review of Ecology and Systematics
25:1–29.
S
TERNER
,R.W.,
AND
K. S
CHULZ
. 1998. Zooplankton
nutrition: recent progress and a reality check.
Aquatic Ecology 32:261–279.
U
RABE
, J., M. K
YLE
,W.M
AKINO
,T.Y
OSHIDA
,T.A
N
-
DERSEN
,
AND
J. J. E
LSER
. 2002. Reduced light in-
creases herbivore production due to stoichiomet-
ric effects of light/nutrient balance. Ecology 83:
619–627.
U
RABE
,J.,
AND
Y. W
ATANABE
. 1992. Possibility of N
limitation or P limitation for planktonic cladoc-
erans: an experimental test. Limnology and
Oceanography 37:244–251.
Received: 27 May 2002
Accepted: 7 October 2002
