Ecological Entomology (2011), 36, 471?479 DOI: 10.1111/j.1365-2311.2011.01290.x
Effects of diet quality on performance and nutrient
regulation in an omnivorous katydid
R A C H E L E . G O E R I Z P E A R S O N,1 S P E N C E R T . B E H M E R ,2 D A N I E L
S . G R U N E R 1 and R O B E R T F . D E N N O 1 1Department of Entomology, University of Maryland,
College Park, Maryland, U.S.A. and 2Department of Entomology, Texas A & M University, College Station, Texas, U.S.A.
Abstract. 1. Omnivores by definition eat both plants and animals. However, little
is known about how diet macronutrient content affects omnivore performance, or the
extent to which they can regulate macronutrient intake. We assessed these questions
using the salt marsh katydid, Conocephalus spartinae Fox (Tettigoniidae).
2. In our first experiment we used artificial diets with different protein?carbohydrate
ratios to assess the effects of diet quality on survival, growth, and lipid accumulation.
We found that diets with a high protein?carbohydrate ratio negatively affected
Conocephalus survival. Among surviving individuals growth was not significantly
different across the treatments, but lipid content decreased significantly as the
protein?carbohydrate ratio of diets increased.
3. In a second experiment we explored the ability of Conocephalus to regulate
their protein?carbohydrate intake. Results revealed that Conocephalus did not feed
randomly when presented with two nutritionally complementary foods. A detailed
analysis of their protein?carbohydrate intake revealed selection for a protein-biased
diet, but a lack of tight regulate of protein?carbohydrate intake.
4. We discuss how key macronutrients can limit omnivores, and how nutri-
tional flexibility may enable omnivores to persist in nutritionally heterogeneous
environments.
Key words. digestable carbohydrates, feeding behoviour, foraging, geometric frame-
work, nutrient regulation, nutrition, omnivore, protein, Tettigoniidae.
Introduction From an omnivore?s perspective, plants and prey are
very different in terms of their nutritional profiles, whether
True omnivores (sensu Coll & Guershon, 2002) consume both considered in terms of elements (e.g. C, N, P, Sterner &
plants and animals, and in terrestrial systems they can influence Elser, 2002) or macronutrients (protein, carbohydrates, and
species and functional diversity (Agrawal & Klein, 2000; lipids, Raubenheimer et al., 2009). For example, N and
Eubanks & Denno, 2000b; Zhi et al., 2006; Ho & Pennings, protein, essential building blocks for structural growth and
2008). As is the case for strict herbivores and predators, reproduction, are typically found in greater quantities in animal
food resource limitations can potentially have large impacts tissue (Matsumura et al., 2004; Taiz & Zeiger, 2006). In
on an omnivore?s fitness (Kaspari et al., 2001; Jacot et al., contrast, digestible carbohydrates and lipids are functionally
2009; Sasakawa, 2009). However, the ability of omnivores to
similar (i.e. they provide energy). However, lipids have greater
consume both plant and animal material provides flexibility in
caloric value (9 vs. 4 kcal g?1, respectively), thus lipids are
adjusting to variable resource supplies. This flexibility could
2.25 times more energy-rich than are digestible carbohydrates.
result in a decreased sensitivity to limitations in food resources.
A key issue, therefore, is how long can an omnivore tolerate Animal tissue typically has greater fat content than plants, but
feeding exclusively on plant material in the absence of prey, plants tend to have greater amounts of simple and complex
or alternatively, feeding only on prey. digestible carbohydrates (e.g. sucrose and starch, respectively)
compared with animal prey items (which contain some simple
Correspondence: Rachel E. Goeriz Pearson, Department of Ento- sugars (e.g. trehalose and glycogen). All organisms, however,
mology, University of Maryland, College Park, MD 20742-4454, require multiple nutrients simultaneously, so it is most useful
U.S.A. E-mail: goerizr@yahoo.com to consider the nutrient balance of foods and the extent to
? 2011 The Authors
Ecological Entomology ? 2011 The Royal Entomological Society 471
472 Rachel E. Goeriz Pearson et al.
which the blend of nutrients in a food best meets a given Methods
organisms nutritional requirements (Raubenheimer & Simpson,
1999; Behmer, 2009). When plants and prey items are viewed Study system
in terms of their position in nutrient space (sensu the Geometric
Framework; Simpson & Raubenheimer, 1999), plants range Early instar nymphs of the omnivorous katydid Cono-
from being energy-biased (rich in digestible carbohydrates cephalus were collected during June 2007 at a study site (a
mid-Atlantic coast intertidal march) located on the Eastern
relative to protein) to having a balanced protein?energy ratios.
shore of Maryland on Chincoteague Bay (38.13?N, 75.30?W).
In contrast, animal prey items tend to have protein?energy
Conocephalus is the most common omnivore at this site,
ratio that range from equal to protein-biased (R.E.G. Pearson
and is recorded as feeding at three different trophic lev-
and R.F. Denno, unpublished; Chapman, 1998; Raubenheimer
els (plant leaves and seeds, herbivores, and predators (Vince
et al., 2009).
et al., 1981; Bertness & Shumway, 1992; Gwynne, 2001). The
When food resources are not limiting, omnivores can mix dominant plants, insect herbivores, and invertebrate predators
their diet to more closely match their nutritional needs. Numer- upon which Conocephalus is likely to feed, and the average
ous studies have shown that insect herbivores (reviewed by macronutrient content of these respective food items, where
Behmer, 2009), and more recently arthropod predators (Mayntz available, are described below.
et al., 2005), actively regulate their nutrient intake, using both The salt marsh cordgrass Spartina alterniflora Loisel
behavioural and physiological mechanisms, to redress nutri- (Poaceae, hereafter Spartina) is the dominant plant in this
tional imbalances. In the case of both insect herbivores and system, and Conocephalus regularly feeds upon it (Denno,
arthropod predators, nutrient regulation directly optimises per- 1983; Gallagher et al., 1988). The N content of Spartina can
formance and fitness (e.g. Mayntz & Toft, 2001; Simpson vary, but at this site it typically ranges between 1% and 2%
et al., 2004; Behmer & Joern, 2008; Toft et al., 2010). How- (Matsumura et al., 2004); using a standard conversion fac-
ever, apart from a single study on an extreme generalist tor of 6.25%, this translates into protein content of between
cockroach, Blatella germanica L. (Blattellidae) (Raubenheimer ?6 and 13%. With respect to carbohydrates and lipids, McIn-
& Jones, 2006), we know very little about the extent to tire and Dunstan (1976), using Spartina plants from Georgia,
which omnivores regulate their macronutrient intake, or how reported that total non-structural (TNS) carbohydrates [free
the macronutrient composition of available foods influence sugars, disaccharides (sucrose), and storage products such as
the fitness of insect omnivores. For example, are omnivores starch, fructose and other oligosaccharides] ranged from 4 to
limited more by protein or energy, or do they require a 10%. Squiers and Good (1974) reported Spartina lipid values
balanced ratio of protein and energy? The nutritional land- of about 2% dry mass. The protein?energy nutrient space that
scape that an omnivore occupies is much broader than that Spartina typically occupies is shown in Fig. 1; here energy is
of a strict herbivore or an arthropod predator (Raubenheimer the summed product of carbohydrates and lipids, but expressed
et al., 2009). Therefore, when resources are in abundant sup-
ply, omnivores will have ample opportunity to optimise their
p30:c50 p40:c40
macronutrient intake by mixing among the available food 80
items.
na
Omnivory is widespread in nature (Coll & Guershon, 2002; 70 ti
pa
r
Thompson et al., 2007), so understanding the factors that S
60
underlie feeding choices in omnivores is fundamental to the
study of population ecology and food-web dynamics (Menge  50 lisia
& Sutherland, 1987; Fagan, 1997; Polis et al., 1997; McCann e p50:c30Pro
k
et al., 1998; Rosenheim, 1998; Eubanks & Denno, 2000a,b), as 40
well as to biological control (Rosenheim et al., 1993; Hodge,
1999). Omnivores have either stabilising or destabilise food 30
ardos
a
web dynamics, depending on the relative strength of their P20
effects on plants and prey (Fagan, 1997; Eubanks & Denno,
2000a,b). Understanding the relative importance and potential 10
interactive effects of the various resources that influence p75:c5
omnivore performance helps to elucidate their effects on 0
food web stability. In this study, we hypothesise that the 0 10 20 30 40 50 60 70 80
omnivorous salt marsh katydid Conocephalus spartinae Fox Protein content (%)
(Tettigoniidae, hereafter Conocephalus) can tolerate a wide
Fig. 1. The protein and energy (carbohydrate equivalent) nutrient
range of nutritional ratios, but performs best on diets that have space of the foods available to Conocephalus. This figure shows marsh
protein?energy content that is protein-biased. Additionally, representative food items from three trophic levels: a plant (Spartina),
we predict that Conocephalus, when allowed to self-select its a herbivore (Prokelisia) and a predator (Pardasa); see the methods for
diet from among nutritionally complementary foods, actively an explanation of carbohydrate equivalents. The dashed lines show the
regulates its protein?energy intake to levels that result in protein-carbohydrate (p:c) ratios of the four experimental foods used
optimal performance. in this study (see the methods for diet making details).
? 2011 The Authors
Ecological Entomology ? 2011 The Royal Entomological Society, Ecological Entomology, 36, 471?479
Carbohydrate equivalents (%)
Diet macronutrient effects on an omnivore 473
as carbohydrate equivalents, e.g. one lipid unit is equal to 2.25 interesting topic, but not addressed in the current study). The
carbohydrate units (as explained in the Introduction), so the remaining 20% of each diet had identical proportions of other
lower energy range of Spartina is about 8.5% carbohydrate ingredients, including vitamins, cholesterol, salts, fatty acids
equivalents [4% from digestible carbohydrates, plus 4.5% from and cellulose (Table S1). We presented the food suspended in
lipids (2% lipid ? 2.25 carbohydrate equivalents); the higher a 1% agar solution, at a 1 : 6 dry diet to agar ratio (Lee et al.,
range is about 14.5% carbohydrate equivalents]. It is also worth 2004). Treatments were replicated 10 times (five individual
noting that Spartina contains plant secondary compounds (par- males and five individual females).
ticularly phenolics). In its southern range phenolics can occur We collected early instar Conocephalus using a sweep net
at concentrations that deter herbivores (Pennings et al., 1998; and kept them for 48 h with only a water source prior to
Salgado & Pennings, 2005); at Spartina?s northern range (our the experiment. Immediately prior to the start of the experi-
study site) phenolics occur at lower concentrations and do not ment, the starting wet-weight mass of each Conocephalus was
deter orthopteran herbivores (Siska et al., 2002; Salgado & measured and individuals were assigned randomly to circular
Pennings, 2005). feeding arenas (15 cm diameter; 6 cm height) that housed four
Hemipteran sap-feeders (planthoppers, leafhoppers, and evenly spaced feeding dishes each containing the same diet and
mirid bugs) are the most abundant insect herbivores at our a water source in the centre. All of the cages were maintained
study site (Denno et al., 1980; Denno, 1983; Denno & Peter- in a growth chamber at a temperature of 27 ?C under a LD
son, 2000), with two phloem-feeding planthoppers, Proke- 13 : 11 h cycle to approximate summer conditions. We sep-
lisia dolus and P. marginata (Delphacidae), the most plentiful
arated the cages with partitions so that Conocephalus could
(Denno et al., 2000). The dry N content of these two planthop-
? not see each other. The experiment was conducted over thepers has been measured at approximately 9%, or about 56%
course of 12 days. Every 2 days we monitored the units for
protein (Matsumura et al., 2004). In contrast to plants, inver-
tebrates (including planthoppers) generally have lower levels Conocephalus survival and replaced food and water sources.
of TNS carbohydrates, and higher levels of lipids. In the case Following the completion of the experiment we weighed
of hemipterans at our site, typical TNS carbohydrate levels the surviving Conocephalus and stored them in a ?20
?C
would be 1?2% (a combination of trehalose glycogen; Becker freezer for further processing. Lipid content of individuals?
et al., 1996), while lipid levels range between 17% and 21% was determined by first drying individuals in a 60 C oven
(Pearson, 2009). When lipids are converted to carbohydrate and weighing them to the nearest 0.001 mg. We extracted
equivalents, the protein?energy content of planthoppers ranges lipids from the dried Conocephalus using a chloroform wash
from equal-ratio to protein-biased (Fig. 1). (Loveridge, 1973). Here Conocephalus were suspended three
Finally, the most abundant invertebrate predators at this site times in succession in a chloroform bath for 24 h. Following
are wolf spiders, particularly Pardosa littoralis (Lycosidae; this procedure they were dried at 60
?C in a drying oven to a
Dobel et al., 1990; Denno et al., 2002). Their dry N content is constant mass and then reweighed. Lipid mass was calculated
estimated at about 12% (Matsumura et al., 2004), giving them as the difference in the two mass measurements.
a protein content of ?75%. Their TNS carbohydrate levels are Analyses for the performance measures of survival and
likely similar to planthoppers, and their lipid levels range from growth proceeded first by anova confirming that the initial
12% to 18% (Salomon et al., 2008; Jensen et al., 2010); the mass of insects did not differ significantly across treatments
likely protein?energy values that Pardosa spiders provide is (SAS: Proc Mixed). Likewise, there was no effect of gender on
shown in Fig. 1. any of the performance variables tested; gender as an effect was
therefore removed from all future analyses. Next we analysed
Conocephalus survival on the different treatments using the
Performance on different diets non-parametric log-rank test (SAS: Proc Lifetest). Data were
In our first experiment we explored how food protein and right-censored to account for the individuals that did not die by
energy content affected Conocephalus survival and growth. the time the experiment ended. To determine if Conocephalus
We did this by creating experimental foods that differed in survival was related to initial mass we performed a post hoc
their protein and digestible carbohydrate content [as outlined in test using anova with mass as the response and survival as a
Dadd (1961), and modified by Simpson and Abisgold (1985)]. factor with two levels (SAS: Proc Mixed).
Although Conocephalus food contains energy in the form of All of the analyses for growth used Conocephalus individ-
both digestible carbohydrates and lipids, for simplicity we pro- uals that survived the duration of the experiment. The best
vided energy only in the form of digestible carbohydrates fitting models for mass gained and lipid mass were selected
(henceforth simply carbohydrates). In total, four diets were with Akaike Information Criterion, which accounts for vari-
generated: (i) p30 : c50 (30% protein and 50% carbohydrate), ance explained and the number of parameters. To examine
(ii) p40 : c40, (iii) p50 : c30, and (iv) p75 : c5. These diets the effect of diet composition on Conocephalus growth we
(with the exception of the heavily biased protein diet) fall analysed total mass gained using ancova (SAS: Proc GLM)
across the protein?energy nutritional space that Conocephalus with diet type as the explanatory variable and initial mass as a
would encounter under natural conditions in the field (Table covariate. We analysed the data for differences in lipid mass of
S1; Fig. 1). It is worth noting that all the foods contain the Conocephalus on the different diets using anova (SAS:
identical macronutrient content (80%); this was done to pre- Proc Mixed) with multiple comparisons using diet type and
vent confounding factors associated with nutrient dilution (an initial mass as the explanatory variables.
? 2011 The Authors
Ecological Entomology ? 2011 The Royal Entomological Society, Ecological Entomology, 36, 471?479
474 Rachel E. Goeriz Pearson et al.
Food and nutrient regulation 100
95
To determine whether omnivorous early instar Conocephalus
nymphs regulate their feeding behaviour, we performed a 90
choice experiment in which individuals were presented with 85
pairs of nutritionally complementary foods. The four diets
80
were the same as those used in the first experiment above
(?Performance on different diets?), except that we assigned 75
three treatments: (i) p30 : c50 paired with p75 : c5, (ii) 70
p40 : c40 paired with p75 : c5, and (iii) p50 : c30 paired with
65 30:50
p75 : c5. Each treatment had 10 replicates (5 males and 5 40:40
females) for a total of 30 experimental units. Conocephalus 60 50:30
75:5
were collected in the field in June 2007 and maintained before 55
and during the experiment in the same manner as above. 50
These experiments were completed over a 6-day period, and 0 2 4 6 8 10 12
the diet cubes in each arena were replaced every 48 h with Time (days)
a fresh cube. Cubes were weighed both before placement
in the arena and at the end of the 48 h feeding period. Fig. 2. Percent survival for each of the diet treatments. Pairwise
comparisons showed a significant difference in the percent survival
To determine the amount of each diet consumed, we first of Conocephalus in the 75:5 treatment group compared to 30:50
assumed that there was minimal loss of water during the group (p = 0.0417), the 40:40 group (p = 0.0315) and the 50:30 group
time that the food was in the cage because of how the (p = 0.0315).
cage was constructed (minimal venting) and the condition of
the food when it was removed. Therefore, from the initial
weight of the cube, the final weight and the amount of water There was a main effect of initial mass on mass gained
(75%) in the diet were subtracted. These data were analysed (F1,24 = 10.07, P = 0.004; Fig. 3a), but there was no signifi-
using t-tests to compare the difference in the mean amount cant treatment effect (F3,24 = 1.54, P = 0.231), or treatment-
of diet consumed in each treatment for three time periods by-initial mass interaction (F3,24 = 1.27, P = 0.306). Analysis
(days 0?2, 2?4, and 4?6), and over the entire experiment of lipids in the Conocephalus carcasses showed a signif-
(days 0?6). icant treatment effect (F3,31 = 14.56, P < 0.0001; Fig. 3b)
Two approaches were employed to determine if the Cono- and a significant effect of initial Conocephalus mass (F1,31 =
cephalus regulated their dietary intake of proteins and car- 22.72, P < 0.0001). In general, lipid body content decreased
bohydrates. First, we scored whether the insects fed ran- as the protein?carbohydrate ratio of the food decreased.
domly. Here a two-tailed, one sample t-test was used to Lipid contents were highest on the p30 : c50 (17.51% ? 1.47)
compare the amounts of the two different foods eaten. and p40 : c40 (15.06% ? 0.87) diets, intermediate on the
Next, we used a manova approach to compare the pro- p50 : c30 diet (12.56% ? 0.96), and lowest on the p75 : c5
tein?carbohydrate intake points for each treatment, with gen- diet (6.64% ? 0.97; Fig. 3b).
der and initial mass as additional explanatory variables (SAS
9.1.2). Again, the amount of water in each of the diets
(75%) was subtracted before analysis was completed. These Food and nutrient regulation
data were analysed for three time periods (days 0?2, 0?4,
When the diets differed significantly in their composi-
and 0?6). For each of the three time periods, the pro-
tion (p30 : c50 paired with p75 : c5; p40 : c40 paired with
tein and carbohydrate consumed met the assumptions of
p75 : c5), Conocephalus preferred the protein-biased diet, but
homogeneity of variance and normality required to perform
when the diet compositions were less divergent (p50 : c30
anova.
paired with p75 : c5), they ate similar amounts of both (Fig. 4).
On the treatment containing p30 : c50 food Conocephalus
Results showed a significant preference for the highly protein-biased
food (p75 : c5) during two time periods (0?2 days and
Performance on different diets 2?4 days), and summed over the entire time period. Indi-
viduals on the treatment containing p40 : c40 food showed a
The log-rank test indicated that survival was poorest on significant preference for the p75 : c5 food during two of the
the highly protein-biased diets (p75 : c5; Fig. 2) and there time periods, and when summed over the entire time period.
was an overall difference in survival between the treatments In contrast, when p50 : c30 and p75 : c5 food was paired, no
over the 12 days of this experiment (?2 = 12.80, P = 0.005). significant preference was detected.
There was a slight decrease in survival on the p50 : c30 diet, Having shown that feeding was not a random process, we
but this difference was not significant compared with sur- next used manova approaches to test whether Conocephalus
vival on the p40 : c40 and p30 : c50 diets. Initial mass at the tightly regulated their protein?carbohydrate intake; if the
start of the experiment did not affect survival (F1,37 = 1.23, intake points for each diet pairing overlapped, we would
P = 0.274). conclude that Conocephalus is capable of tight nutrient
? 2011 The Authors
Ecological Entomology ? 2011 The Royal Entomological Society, Ecological Entomology, 36, 471?479
Survival (%)
Diet macronutrient effects on an omnivore 475
(a) 4.0 more closely with a recent paper by Raubenheimer et al. (2009)
that suggests predators may in fact be energy limited, because
3.5 their diets are more likely to be deficient in carbohydrates or
3.0 lipids. Recent empirical work by Hawlena and Schmitz (2010)
suggests that energy limitation may depend on the context: the
2.5 stress of predation risk increased carbohydrate metabolism in
orthopteran herbivores. In the field, under natural conditions,
2.0
Conocephalus consumes a mixed diet of plants (leaves, pollen,
1.5 and seeds; Bertness & Shumway, 1992; Sala et al., 2008) and
animal prey [confirmed by Vince et al. (1981) who found insect
1.0 remains in their frass]. However, gut content analysis of field-
0.5 caught individuals (Pearson, 2009) suggests that they may be
more predator than vegetarian; 88% of the collected individuals
0.0 contained arthropod remains, but only 54% contained plant
matter. Eating a diet rich in invertebrates should provide a
(b) 20 a sufficient amount of protein (see Fig. 1), but if invertebrate
ab prey items are lean (i.e. low in energy), feeding on plant
16
b material (e.g. Spartina leaves) may be an effective strategy for
Conocephalus to redress energy deficiencies, because generally
12 Spartina has a lower protein?energy ratio compared with
c animal prey (see Fig. 1). The other advantage to feeding
8 on Spartina is that it simultaneously limits over-ingestion
of protein (especially when measured in terms of absolute
4 amounts), and in some instances too much protein may become
toxic (see Raubenheimer & Simpson, 1999; Simpson et al.,
0 2004; Raubenheimer et al., 2005).
30:50 40:40 50:30 75:5 In the short-term Conocephalus may have some capacity
to post-ingestively regulate protein intake, but animals that
Food protein-carbohydrate ratio overeat nutrients in excess of requirements for an extended
Fig. 3. A) Conocephalus mean proportion mass gain (? SE) in the period of time often suffer reduced performance (e.g. Simpson
four diet treatments. Mean proportion gain was calculated as the final et al., 2004; Raubenheimer et al., 2005). Interestingly, another
Conocephalus mass/initial Conocephalus mass; B) Comparison of the omnivore, the German cockroach (Blatella germanica L.), pro-
mean percent body lipids (? SE) of the Conocephalus carcasses fed vides an exception to this general overeating rule. Rauben-
one of four diets differing in protein carbohydrate (P:C) diet treatment heimer and Jones (2006) showed that, across a broad range
for twelve days. Means with different letters are significantly different of protein?carbohydrate concentrations, its survival was unaf-
(P < 0.05). fected (Raubenheimer & Jones, 2006). Here the authors pos-
tulated that it adjusted to variations in the balance of ingested
regulation (e.g. Raubenheimer & Jones, 2006). We observed a nutrients using physiological mechanisms associated with sur-
significant treatment effect for each of the 2-day periods (Day viving long periods of famine. Cockroaches are opportunistic
2: F = 10.03, P < 0.0001; Day 4: F = 11.59, P < scavengers and extreme generalists that have the ability to4,54 4,54
0.0001; Day 6: F4,54 = 13.24, P < 0.0001; Fig. 5). Paired use their fat body to store N, in the form of uric acid, and
contrasts were then used to explore this outcome in greater carbohydrates, in the form of lipids (Douglas, 1989). Cock-
detail, and here we found significant differences in the total roaches also have a number of paunches in the hindgut that
amount of protein and carbohydrate eaten for each of the three house bacteria, which provide essential nutrients (Bourtzis &
pairs (p30 : c50 vs. p40 : c40, p30 : c50 vs. p50 : c30 and Miller, 2003). In contrast, Conocephalus are leaner than cock-
p40 : c40 vs. p50 : c30), and at each of the three time periods. roaches with a less well developed fat body system and with no
Gender did not significantly affect protein?carbohydrate nutritional endosymbionts isolated from their alimentary canals
intake, nor was there an effect of initial mass or an interactive (Nation, 2001).
effect between treatment and initial mass. Despite survival differences among treatments, no differ-
ences in growth were observed for individual Conocephalus
that survived to the end of the 12-day experiment. Analy-
Discussion sis of body lipid content revealed, however, that the pro-
tein?carbohydrate ratio of the diet significantly affected Cono-
Omnivores are important players in terrestrial food webs (e.g. cephalus body composition. If growth is viewed in terms of
Fagan, 1997; Eubanks & Denno, 2000b) but currently we know lean mass (total mass ? body fat), which places an emphasis on
very little about what constitutes an optimal diet for them. structural growth, the best diet for Conocephalus, in terms of
Denno and Fagan (2003) suggested that omnivores in terrestrial combining survival and relative lean mass (total mass ? lipid
systems are likely to be protein limited, but our findings accord content), is the slightly protein-biased one (p50 : c30). That
? 2011 The Authors
Ecological Entomology ? 2011 The Royal Entomological Society, Ecological Entomology, 36, 471?479
Body lipid content (%) Mass gain (as a proportion)
476 Rachel E. Goeriz Pearson et al.
? 2011 The Authors
Ecological Entomology ? 2011 The Royal Entomological Society, Ecological Entomology, 36, 471?479
250 Day 0-Day 2 250 Day 2-Day 4 250 Day 4-Day 6 700 Day 0-Day 6
* * 600200 *200 200
500
150 150 150 400
100 100 100 300
200
50 50 50
100
0 0 0 0
p30:c50 p75:c5 p30:c50 p75:c5 p30:c50 p75:c5 p30:c50 p75:c5
250 250
* *
250 700
* *
600
200 200 200
500
150 150 150 400
100 100 100 300
200
50 50 50
100
0 0 0 0
p40:c40 p75:c5 p40:c40 p75:c5 p40:c40 p75:c5 p40:c40 p75:c5
250 250 250
700
200 200 200 600
500
150 150 150
400
100 100 100 300
200
50 50 50
100
0 0 0 0
p50:c30 p75:c5 p50:c30 p75:c5 p50:c30 p75:c5 p50:c30 p75:c5
Diet Diet Diet Diet
Fig. 4. Comparison of the mean diet eaten (? SE) for each of the four time periods (days 0?2, days 2?4, days 4?6 and days 0?6). Differences (P < 0.05) in the amounts consumed are denoted
by (*).
A m o u n t  e a t e n  ( m g )  A m o u n t  e a t e n  ( m g ) A m o u n t  e a t e n  ( m g )
Diet macronutrient effects on an omnivore 477
30:50 40:40 regulate their protein?energy intake, as has been observed200
repeatedly in insect herbivores (reviewed by Behmer, 2009).
30:50 and 75:5
40:40 and 75:5 Post hoc contrasts revealed that for each time period, none
50:30 and 75:5 of the self-selected intake targets converged on a single
point. Perhaps tight macronutrient regulation isn?t essential
150
if Conocephalus possess efficient post-ingestive physiological
processes that allow compensation for differences in nutri-
50:30 ent intake. Raubenheimer and Simpson (1993) showed that
locusts, over the short-term (a single developmental stadium),
100
could reach similar growth targets across a wide-range of pro-
tein?carbohydrate diets. In cases where carbohydrates are in
excess, post-ingestive processes include respiring excess carbo-
hydrates (Zanotto et al., 1993, 1997) and/or converting carbo-
50
hydrates to fat and storing them (Simpson et al., 2002). Some
insects can process excess N; for example, some orthopter-
75:5 ans (e.g. Locusta migratoria L.) can metabolise excess pro-
0 tein and use amino acids as a source of energy via deam-
0 50 100 150 200 ination (Raubenheimer & Simpson, 2003). It appears that
Protein consumed (mg) Conocephalus can at least convert excess carbohydrates to
fat, but we currently know little about the extent to which
Fig. 5. Bivariate means of protein and carbohydrates consumed (? invertebrate predators or omnivores can produce energy via
95% CL) by Conocephalus spartinae when given one of three paired deamination. Alternatively, if excess protein cannot be used to
diets: () 30:50 and 75:5, () 40:40 and 75:5 and () 50:30 and 75:5. generate energy, protein can be metabolised during digestion,
The first set of points in the series represents the amount consumed
and amino acids in excess of requirements can be voided during
after the first 2 days. The second set of points is the total amount
consumed after 4 days and the final set of points is the total amount excretion (Zanotto et al., 1993). However, the high mortality
consumed after 6 days. The dark lines represent the nutritional rails of of Conocephalus on the heavily protein-biased diets suggests
the four diets as listed in the margin. this is likely not an option.
As an omnivore, Conocephalus includes both plant and
animal material in its diet despite large differences in macronu-
this particular diet is the best in the no-choice experiment for trient composition among dietary resources. In our short-term
Conocephalus is interesting on two fronts. First, it has a similar feeding experiments, Conocephalus flexibly accepted a wide
protein?energy ratio to one of the most abundant prey items range of artificial diets to maintain growth. However, our high
(planthoppers; see Fig. 1). Second, it is close (in terms of its protein, very low energy diet (p75 : c5) clearly shows that car-
nutrient-space position) to the protein?carbohydrate ratio that bohydrates (energy) can be limiting for Conocephalus. Perhaps
was self-selected in the choice experiment. when energy is limiting (either because prey items are lean,
In the field Conocephalus can access a broad range of food or prey items are scarce), plants provide a readily available
types, with different nutrient compositions, and this at the min- energy source, mostly in the form of digestible carbohydrates
imum provides the opportunity to regulate their macronutrient (simple sugars and starch). The apparent short-term nutritional
intake. Results from our choice experiments reveal two key flexibility of Conocephalus makes sense in the resource land-
findings. First, Conocephalus is not a random feeder with scape found on the mid-Atlantic salt marsh where planthopper
respect to its macronutrient intake. We presented Conocephalus prey can reach outbreak numbers and plant quality is variable
nymphs with three different food pairings, and when the bal- over space and time. Conocephalus can utilise these variable
anced (40 : 40) or carbohydrate-biased (30 : 50) foods were conditions by consuming available resources and, like gen-
paired with the highly protein-biased (75 : 5) food, a clear eralist herbivores, consuming unbalanced foods when they are
preference for the protein-rich food was observed over the encountered because the probability of encountering a comple-
entire experiment, and for each time period except the final one mentary food will be high (Raubenheimer & Simpson, 2003;
(days 4?6) where p40 : c40 was present. In contrast, no food Behmer, 2009).
preference was observed when p50 : c30 food was paired with For omnivores like Conocephalus, feeding on plants or
p75 : c5 food. However, this does not rule out the possibility prey may not represent an ?either/or? situation, but rather
of nutrient regulation was occurring; here the protein?energy a nutritional continuum where they benefit by sampling
intake target would have been achieved by eating equally from their surroundings or feeding complementarily to meet their
among the two available food types. Nutrient regulation has nutritional requirements. The degree to which an omnivore
been repeatedly observed in insect herbivores (reviewed by mixes its diet likely depends not on whether a plant or
Behmer, 2009) and demonstrated in predaceous arthropods an animal itself is more nutritious, but rather how each,
(Mayntz et al., 2005; Pekar et al., 2010), and we can now add when combined, fulfil an omnivore?s nutritional needs at that
a non-cockroach insect omnivore to the list. particular time. These needs, of course, can change over
The second key finding was that while Conocephalus time depending on both the state of the omnivore (age, sex
nymphs self-selected a protein-rich diet, they did not tightly etc.) and the state of its environment (resource availability,
? 2011 The Authors
Ecological Entomology ? 2011 The Royal Entomological Society, Ecological Entomology, 36, 471?479
Carbohydrate consumed (mg)
478 Rachel E. Goeriz Pearson et al.
abiotic conditions, toxins, etc.). We propose that to more fully Denno, R.F. (1983) Tracking variable host plants in space and time.
understand how omnivores affect population ecology, food- Variable Plants and Herbivores in Natural and Managed Systems
web dynamics, and biological control, it will be important (ed. by R. F. Denno and M. S. McClure), pp. 291?341. Academic
to gain a better understanding of the functional significance Press, New York, New York.
of switching between plants and prey, how and the extent to Denno, R.F. & Fagan, W.F. (2003) Might nitrogen limitation promote
which they regulate their food intake, and how in turn this omnivory among carnivorous arthropods? Ecology, 84, 2522?2531.
regulation affects their fitness. Denno, R.F. & Peterson, M.A. (2000) Caught between the devil and
the deep blue sea: mobile planthoppers elude natural enemies and
deteriorating host plants. American Entomologist, 46, 95?100.
Acknowledgements Denno, R.F., Raupp, M.J., Tallamy, D.W. & Reichelderfer, C.F.
(1980) Migration in heterogeneous environments: differences in
We thank T. Pearson, D. Lewis, and R. Lewis for helpful com- habitat selection between the wing forms of the dimorphic plan-
ments on versions of the manuscript. We thank R. Pearson?s thopper, Prokelisia marginata (Homoptera: Delphacidae). Ecology,
committee, J. Dietz, C. Mitter, M. Raupp, and G. Wimp, for 61, 859?867.
support and helpful comments. This work was supported by Denno, R.F., Peterson, M.A., Gratton, C., Cheng, J.A., Langellotto,
National Science Foundation Grant DEB-0638813 to R.F.D. G.A., Huberty, A.F. et al. (2000) Feeding-induced changes in
plant quality mediate interspecific competition between sap-feeding
and a Gahan Fellowship from the Department of Entomology,
herbivores. Ecology, 81, 1814?1827.
University of Maryland to R.E.G.P. Denno, R.F., Gratton, C., Peterson, M.A., Langellotto, G.A., Finke,
D.L. & Huberty, A. (2002) Bottom-up forces mediate natural-
Supporting Information enemy impact in a phytophagous insect community. Ecology, 83,
1443?1458.
Additional Supporting Information may be found in the online Dobel, H.G., Denno, R.F. & Coddington, J.A. (1990) Spider (araneae)
version of this article under the DOI reference: community structure in an intertidal salt-marsh ? effects of
vegetation structure and tidal flooding. Environmental Entomology,
10.1111/j.1365-2311.2011.01290.x
19, 1356?1370.
Table S1. Constituents of four artificial diets fed to Douglas, A.E. (1989) Mycetocyte symbiosis in insects. Biological
Conocephalus. Reviews of the Cambridge Philosophical Society, 64, 409?434.
Please note: Neither the Editors nor Wiley-Blackwell Eubanks, M.D. & Denno, R.F. (2000a) Health food versus fast food:
are responsible for the content or functionality of any the effects of prey quality and mobility on prey selection by a
supplementary material supplied by the authors. Any queries generalist predator and indirect interactions among prey species.
(other than missing material) should be directed to the Ecological Entomology, 25, 140?146.
corresponding author for the article. Eubanks, M.D. & Denno, R.F. (2000b) Host plants mediate omnivore-
herbivore interactions and influence prey suppression. Ecology, 81,
936?947.
References Fagan, W.F. (1997) Omnivory as a stabilizing feature of natural
communities. American Naturalist, 150, 554?567.
Agrawal, A.A. & Klein, C.N. (2000) What omnivores eat: direct Gallagher, J.L., Somers, G.F., Grant, D.M. & Seliskar, D.M. (1988)
effects of induced plant resistance on herbivores and indirect Persistent differences in 2 forms of Spartina alterniflora ? a
consequences for diet selection by omnivores. Journal of Animal common garden experiment. Ecology, 69, 1005?1008.
Ecology, 69, 525?535. Gwynne, D.T. (2001) Katydids and Bush-Crickets: Reproductive
Becker, A., Schloder, P., Steele, J.E. & Wegener, G. (1996) The Behavior and Evolution of the Tetigoniidae. Cornell University
regulation of trehalose metabolism in insects. Experientia, 52, Press, Ithaca, New York.
433?439. Hawlena, D. & Schmitz, O.J. (2010) Physiological stress as a fun-
Behmer, S.T. (2009) Insect herbivore nutrient regulation. Annual damental mechanism linking predation to ecosystem functioning.
Review of Entomology, 54, 165?187. American Naturalist, 176, 537?556.
Behmer, S.T. & Joern, A. (2008) Coexisting generalist herbivores Ho, C.K. & Pennings, S.C. (2008) Consequences of omnivory for
occupy unique nutritional feeding niches. Proceedings of the trophic interactions on a salt marsh shrub. Ecology, 89, 1714?1722.
National Academy of Sciences of the United States of America, 105,
Hodge, M.A. (1999) The implications of intraguild predation for the
1977?1982.
role of spiders in biological control. Journal of Arachnology, 27,
Bertness, M.D. & Shumway, S.W. (1992) Consumer driven pollen
limitation of seed production in marsh grasses. 351?362.American Journal
of Botany, 79, 288?293. Jacot, A., Valcu, M., van Oers, K. & Kempenaers, B. (2009) Exper-
Bourtzis, K. & Miller, T.A. (eds) (2003) Insect Symbiosis. CRC Press, imental nest site limitation affects reproductive strategies and
Boca Raton, Florida. parental investment in a hole-nesting passerine. Animal Behaviour,
Chapman, R.F. (1998) The Insects: Structure and Function. Cambridge 77, 1075?1083.
University Press, Cambridge, U.K. Jensen, K., Mayntz, D., Wang, T., Simpson, S.J. & Overgaard, J.
Coll, M. & Guershon, M. (2002) Omnivory in terrestrial arthropods: (2010) Metabolic consequences of feeding and fasting on nutrition-
mixing plant and prey diets. Annual Review of Entomology, 47, ally different diets in the wolf spider Pardosa prativaga. Journal of
267?297. Insect Physiology, 56, 1095?1100.
Dadd, R.H. (1961) The nutritional requirements of locusts. IV. Kaspari, M., Pickering, J. & Windsor, D. (2001) The reproductive
Requirements for vitamins of the B complex. Journal of Insect flight phenology of a neotropical ant assemblage. Ecological
Physiology, 6, 1?12. Entomology, 26, 245?257.
? 2011 The Authors
Ecological Entomology ? 2011 The Royal Entomological Society, Ecological Entomology, 36, 471?479
Diet macronutrient effects on an omnivore 479
Lee, K.P., Raubenheimer, D. & Simpson, S.J. (2004) The effects Sala, N.M., Bertness, M.D. & Silliman, B.R. (2008) The dynamics
of nutritional imbalance on compensatory feeding for cellulose- of bottom-up and top-down control in a New England salt marsh.
mediated dietary dilution in a generalist caterpillar. Physiological Oikos, 117, 1050?1056.
Entomology, 29, 108?117. Salgado, C.S. & Pennings, S.C. (2005) Latitudinal variation in palata-
Loveridge, J.P. (1973) Age and the changes in water and fat content of bility of salt-marsh plants: are differences constitutive? Ecology, 86,
adult laboratory-reared Locausta migratoria migratorioides R. and 1571?1579.
F. Rhodesian Journal of Agricultural Research, 11, 131?143. Salomon, M., Mayntz, D. & Lubin, Y. (2008) Colony nutrition skews
Matsumura, M., Trafelet-Smith, G.M., Gratton, C., Finke, D.L., Fagan, reproduction in a social spider. Behavioral Ecology, 19, 605?611.
W.F. & Denno, R.F. (2004) Does intraguild predation enhance Sasakawa, K. (2009) Effects of maternal diet on fecundity and
predator performance? A stoichiometric perspective. Ecology, 85, larval development in the ?primitive? granivorous carabid Amara
2601?2615. (Curtonotus) macronota. Entomologia Experimentalis et Applicata,
Mayntz, D. & Toft, S. (2001) Nutrient composition of the prey?s diet 130, 106?112.
affects growth and survivorship of a generalist predator. Oecologia, Simpson, S.J. & Abisgold, J.D. (1985) Compensation by locusts for
127, 207?213. changes in dietary nutrients: behavioural mechanisms. Physiological
Mayntz, D., Raubenheimer, D., Salomon, M., Toft, S. & Simpson, S.J. Entomology, 10, 443?452.
(2005) Nutrient-specific foraging in invertebrate predators. Science, Simpson, S.J. & Raubenheimer, D. (1999) Assuaging nutritional
307, 111?113. complexity: a geometrical approach. Proceedings of the Nutrition
McCann, K., Hastings, A. & Huxel, G.R. (1998) Weak trophic inter- Society, 58, 779?789.
actions and the balance of nature. Nature, 395, 794?798. Simpson, S.J., Raubenheimer, D., Behmer, S.T., Whitworth, A. &
McIntire, G.L. & Dunstan, W.M. (1976) Non-structural carbohydrates Wright, G.A. (2002) A comparison of nutritional regulation in
in Spartina alterniflora Loisel. Botanica Marina, 19, 93?96. solitarious- and greagarious-phase nymphs of the desert locust Schis-
Menge, B. & Sutherland, W.J. (1987) Community regulation: variation tocera gregaria. Journal of Experimental Biology, 205, 121?129.
in disturbance, competition and predation in relation to environmen- Simpson, S.J., Sibly, R.M., Lee, K.P., Behmer, S.T. & Rauben-
tal stress and recruitment. American Naturalist, 130, 563?576. heimer, D. (2004) Optimal foraging when regulating intake of mul-
Nation, J.L. (2001) Insect Physiology and Biochemistry. CRC Press, tiple nutrients. Animal Behaviour, 68, 1299?1311.
Boca Raton, Florida. Siska, E.L., Pennings, S.C., Buck, T.L. & Hanisak, M.D. (2002)
Pearson, R.E.G. (2009) Nutrient regulation by an omnivore and the Latitudinal variation in palatability of salt-marsh plants: which traits
effects on performance and distribution. Dissertation, University of are responsible? Ecology, 83, 3369?3381.
Maryland, College Park, Maryland. Squiers, E.R. & Good, R.E. (1974) Seasonal changes in the produc-
Pekar, S., Mayntz, D., Ribeiro, T. & Herberstein, M.E. (2010) Spe- tivity, caloric content, and chemical composition of a population of
cialist ant-eating spiders selectively feed on different body parts to salt-marsh cord-grass (Spartina alterniflora). Chesapeake Science,
balance nutrient intake. Animal Behaviour, 79, 1301?1306. 15, 63?71.
Pennings, S.C., Carefoot, T.H., Siska, E.L., Chase, M.E. & Page, T.A. Sterner, R.W. & Elser, J.J. (2002) Ecological Stoichiometry the
(1998) Feeding preferences of a generalist salt-marsh crab: relative Biology of Elements from Molecules to the Biosphere. Princeton
importance of multiple plant traits. Ecology, 79, 1968?1979. University Press, Princeton, New Jersey.
Polis, G.A., Anderson, W.B. & Holt, R.D. (1997) Toward an integra- Taiz, L. & Zeiger, E. (2006) Plant Physiology, Sinauer Associates,
tion of landscape and food web ecology: the dynamics of spatially Sunderland, Massachusetts.
subsidized food webs. Annual Review of Ecology and Systematics, Thompson, R.M., Hemberg, M., Starzomski, B.M. & Shurin, J.B.
28, 289?316. (2007) Trophic levels and trophic tangles: the prevalence of
Raubenheimer, D. & Jones, S.A. (2006) Nutritional imbalance in omnivory in real food webs. Ecology, 88, 612?617.
an extreme generalist omnivore: tolerance and recovery through Toft, S., Li, D. & Mayntz, D. (2010) A specialized araneophagic
complementary food selection. Animal Behaviour, 71, 1253?1262. predator?s short-term nutrient utilization depends on the macronu-
Raubenheimer, D. & Simpson, S.J. (1993) The geometry of compen- trient content of prey rather than on prey taxonomic affiliation.
satory feeding in the locust. Animal Behaviour, 45, 953?964. Physiological Entomology, 35, 317?327.
Raubenheimer, D. & Simpson, S.J. (1999) Integrating nutrition: a Vince, S.W., Valiela, I. & Teal, J.M. (1981) An experimental-study
geometrical approach. Entomolgia Experimentalis et Applicata, 91, of the structure of herbivorous insect communities in a salt-marsh.
67?82. Ecology, 62, 1662?1678.
Raubenheimer, D. & Simpson, S.J. (2003) Nutrient balancing in Zanotto, F.P., Simpson, S.J. & Raubenheimer, D. (1993) The regula-
grasshoppers: behavioural and physiological correlates of dietary tion of growth by locusts through postingestive compensation for
breadth. Journal of Experimental Biology, 206, 1669?1681. variation in the levels of dietary-protein and carbohydrate. Physio-
Raubenheimer, D., Lee, K.P. & Simpson, S.J. (2005) Does Bertrand?s logical Entomology, 18, 425?434.
rule apply to macronutrients? Proceedings of the Royal Society of Zanotto, F.P., Gouveia, S.M., Simpson, S.J., Raubenheimer, D. &
London Series B, 272, 2429?2434. Calder, P.C. (1997) Nutritional homeostasis in locusts: is there a
Raubenheimer, D., Simpson, S.J. & Mayntz, D. (2009) Nutrition, mechanism for increased energy expenditure during carbohydrate
ecology and nutritional ecology: toward an integrated framework. overfeeding? Journal of Experimental Biology, 200, 2437?2448.
Functional Ecology, 23, 4?16. Zhi, J.R., Margolies, D.C., Nechols, J.R. & Boyer, J.E. (2006) Host-
Rosenheim, J.A. (1998) Higher-order predators and the regulation of plant-mediated interaction between populations of a true omnivore
insect populations. Annual Review of Entomology, 43, 421?447. and its herbivorous prey. Entomologia Experimentalis et Applicata,
Rosenheim, J.A., Wilhoit, L.R. & Armer, C.A. (1993) Influence of 121, 59?66.
intraguild predation among generalist insect predators on the
suppression of an herbivore population. Oecologia, 96, 439?449. Accepted 20 April 2011
? 2011 The Authors
Ecological Entomology ? 2011 The Royal Entomological Society, Ecological Entomology, 36, 471?479