SEARCH

SEARCH BY CITATION

Keywords:

  • Carbon;
  • nutrient balance hypothesis;
  • condensed tannin;
  • cyanogenic glycoside;
  • herbivory;
  • Lycaenidae;
  • Polyommatus icarus

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Abstract 1. Plant growth and chemical defence compounds in four Lotus corniculatus genotypes exposed to factorial combinations of ambient and elevated carbon dioxide, and herbivory by caterpillars of Polyommatus icarus were measured to test the predictions of the carbon/nutrient balance hypothesis.

2. Shoot and root biomass, allocation to shoots versus roots, and carbon-based defence compounds were greater under elevated carbon dioxide. Pupal weight of P. icarus was greater and development time shorter under elevated carbon dioxide.

3. Herbivory decreased shoot growth relative to root growth and production of nitrogen-based defence (cyanide). Young leaves contained more defence compounds than old leaves, and this response depended on carbon dioxide and herbivory treatments (significant interactions).

4. Genotype-specific responses of plants to carbon dioxide and herbivory were found for the production of cyanide. Furthermore, maternal butterfly-specific responses of caterpillars to carbon dioxide were found for development time. This suggests the existence of genetic variation for important defence and life-history traits in plants and herbivores in response to rising carbon dioxide levels.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Carbon cycle models project atmospheric carbon dioxide (CO2) concentrations of 540–970 p.p.m. by the end of the century as a result of fossil fuel consumption and global deforestation (e.g. Houghton et al., 1996; www.ipcc.ch). Enriched atmospheric CO2 can increase plant growth and alter leaf tissue chemistry, thereby altering trophic interactions, such as those between plants and herbivores. Changes in leaf chemistry under elevated CO2, especially reduced leaf protein concentration, could lead to decreased insect herbivore growth, increased larval development time, or compensatory feeding (Fajer et al., 1989; Ayres, 1993; Lincoln et al., 1993; Watt et al., 1995). Changes in leaf nitrogen should have the greatest impact on herbivory because nitrogen is most limiting for insect growth (Mattson, 1980).

The impact of CO2 enrichment on herbivores has been investigated in about 60 studies (e.g. Ayres, 1993; Lincoln et al., 1993; Watt et al., 1995; Bezemer & Jones, 1998; Coviella & Trumble, 1999; Whittaker, 1999; Goverde et al., 2002). In contrast, little is known about plant response to herbivory under enriched CO2 atmospheres and only one study has used real rather than simulated herbivory (Awmack & Harrington, 2000).

Many plants respond to defoliation with compensatory growth and induced response (e.g. Bryant et al., 1991; Karban, 1993; Karban & Baldwin, 1997). Elevated CO2 could influence this plant response because higher concentrations of CO2 might have a positive effect on plant growth as well as a plant's ability to produce conservative or induced secondary compounds as protection against herbivores. Simulated defoliation causes species-specific reaction in terms of compensatory growth. Some species compensate better or completely at elevated CO2 (Ryle & Powell, 1992; Kruger et al., 1998), some better at ambient CO2 (Lovelock et al., 1999), or equally in both environments (Fajer et al., 1991; Wilsey et al., 1994, 1997; Kruger et al., 1998). Furthermore, induced responses in plants may influence insect performance (Rhoades, 1983; Karban & Baldwin, 1997). For example, larval performance and preference of Lymantria dispar depend on prior plant injury, which induces changes in foliar chemistry (Rieske & Raffa, 1998; Havill & Raffa, 1999), and prior herbivory on Festuca arundinacea has a negative effect on larval mass, development time, and assimilation efficiency of Spodoptera frugiperda (Bultman & Conard, 1998).

The carbon/nutrient balance hypothesis (Bryant et al., 1983; Coley et al., 1985; Tuomi et al., 1988, 1990; Herms & Mattson, 1992; Koricheva et al., 1998) proposes that induced chemical resistance and regrowth after defoliation are affected by a plant's carbon/nutrient balance. Under low carbon/nutrient environmental conditions (e.g. low light, low CO2, high nitrogen), shoot growth should take priority over root growth, and the mobilisation of carbohydrates for this activity should lead to a decrease in leaf carbon-based defence compounds relative to nitrogen-based compounds. Under high carbon/nutrient conditions (e.g. high CO2, low nitrogen), carbon will be more readily available and this should attenuate the effects of defoliation on root/shoot partitioning and on the shifts in the relative abundance of carbon- and nitrogen-rich defence compounds. Further models predict allocation to secondary compounds depending on source and sink strength (growth-differentiation model; Herms & Mattson, 1992) or their dependence on phenylalanine (Protein Competition Model; Jones & Hartley, 1999; Bezemer et al., 2000).

Studies of the effects of elevated CO2 and nutrient availability on plant defence molecules generally support the carbon/ nutrient balance hypothesis (Fajer et al., 1992; Lindroth et al., 1995; Kinney et al., 1997; Koricheva et al., 1998; Penuelas & Estiarte, 1998), however detection of such environmental effects on defence compounds is complicated because their concentrations also change with leaf ontogeny (Raupp & Denno, 1983; Jones, 1988; Gleadow et al., 1998; Williams et al., 1998; Gleadow & Woodrow, 2000a) and differ among genotypes (Briggs, 1990; Briggs & Schultz, 1990; Fajer et al., 1992).

The work reported here was designed to test the influence of elevated CO2 on plant response to herbivory in birdsfoot trefoil Lotus corniculatus L., which has two types of defence molecules: condensed tannins (carbon-based defence compounds) and cyanogenic glycosides (nitrogen-based defence compounds). Lotus corniculatus shows responses to nitrogen and CO2 that are not always in concordance with the carbon/nutrient balance hypothesis (Briggs, 1990; Briggs & Schultz, 1990; A. Bazin, unpublished). On the other hand, grazing by caterpillars of the moth Spodoptera eridania induces condensed tannin production and compensatory shoot growth (Briggs, 1991).

The work compared the defence molecule concentrations for plants exposed or not exposed to herbivores in old and young leaves of four L. corniculatus genotypes that differ in the constitution of their defence levels (Goverde et al., 1999). In addition, insect performance on these four genotypes under ambient and elevated CO2 conditions is reported. The following predictions were made. (1) Under elevated CO2, plants will grow more, increase partitioning to roots, reduce leaf cyanogenic glycoside concentrations, and increase concentrations of total polyphenols and condensed tannins. (2) Following defoliation at elevated CO2, faster compensatory regrowth will occur due to increased carbon reserves. (3) Herbivory will accentuate differences in concentrations of carbon-based defence molecules between ambient and elevated CO2 treatments by accentuating the trade-off between utilisation of carbon reserves for growth vs. production of defence compounds.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Plant material and growth conditions

Lotus corniculatus L. plants were collected in a meadow south of Paris, France (48°42′08′N, 2°10′17′E). Plants were planted in pots filled with compost, and cultivated in a greenhouse at the University of Paris, Orsay. Four genotypes, chosen for their difference in condensed tannin and cyanide concentrations, were cloned from shoot cuttings made on 21 July 1997. On 15 August, following the development of healthy root systems, cuttings were transplanted to 1-litre plastic pots containing a mixture of 1:1 compost and soil from calcareous grassland in the Swiss Jura mountains (Vicques, 570 m a.s.l.). For each of the four genotypes, 17 plants were assigned randomly to each of the two CO2 treatments: 350 p.p.m. (ambient) or 700 p.p.m. (elevated) CO2. Thus a total of 136 plants was used in the experiment (17 individuals × four genotypes × two CO2 treatments). Because only one growth chamber per treatment was used, pot position was re-randomised within each chamber and plants were moved from one chamber to the other, reversing the CO2 treatment between chambers every week, to limit chamber effects. Each pot was used as a replicate. Plants were watered when necessary and fertilised once a week using a commercial nutrient solution (N: 117.5, P: 40.0, K: 134.0 mg L−1). Temperatures were 24 °C day and 15 °C night with natural light plus sodium lights for a 16-h photoperiod.

Butterfly rearing

Females of Polyommatus icarus Rottemburg were captured at two sites in the Swiss Jura mountains (Nenzlingen 47°27′05′N, 7°33′80′E; Liesberg 47°24′30′N, 7°25′05′E). Eggs were obtained from four different females (for oviposition details, see Goverde et al., 1999). On 10–12 September 1997, batches of eight to 10 caterpillars from each of the four females were placed in Petri dishes (9 cm diameter). These dishes were assigned across the four plant genotypes and the two CO2 treatments. Petri dishes contained a moistened piece of paper towel (15 × 50 mm) to maintain air humidity and were covered with nylon mesh and the Petri dish lid. Petri dishes were placed in a growth chamber at 25 °C, a LD 16:8 h cycle, and ambient CO2 conditions. Every second day, a plant from the appropriate genotype and CO2 treatment was chosen, fresh 10–15 cm long sprigs were cut, and the stems were placed into a 1.5-ml EppendorfTM tube full of water with a pierced cap, and given to the larvae. Larvae were batch-raised in this manner for the first two instars, during which they are very small and difficult to manipulate individually without great risk of injury. Third-instar larvae were then used for the feeding trial.

Feeding trial

On 17–19 September 1997, freshly moulted third-instar larvae were individually weighed and transferred to the plants in the greenhouse. From each plant genotype, 12 plants per CO2 treatment received each one caterpillar, three plants each for the four females. Five control plants per genotype and CO2 treatment received no caterpillar. All treatments and control plants were enclosed in cages of mellinex foil (9.5 × 9.5 × 30 cm) with 8-cm diameter windows on each side covered by nylon mesh and a pierced cover to allow gas exchange.

Twenty-four to 48 h after pupation, each individual was weighed and placed in one of the 18 compartments of a plastic box (210 × 120 mm) with a nylon mesh floor. Boxes were placed 2 cm above a pan of water in the same growth chambers as the plants, and were checked twice a day (08.00 and 20.00 hours) for emerged butterflies. After they emptied their gut, emerged butterflies were freeze killed, sexed, dried at 65 °C for 48 h, and weighed.

Foliar chemistry

Just before pupation, samples of young and old leaves were collected from all plants for cyanide and polyphenol analysis. Old, fully expanded leaves that had developed before the beginning of the herbivory treatment were collected from all herbivory treatment and control plants. In the herbivory treatment, no young leaves were available on some plants because the caterpillars had consumed all young tissue on them completely, so samples were obtained from only 65 of these 96 plants. After pupation, the remaining plant material was harvested, separated into roots and shoots, dried at 38 °C for 48 h, and weighed.

For total polyphenol and condensed tannin analysis, dried leaf samples were ground with liquid nitrogen and quartz sand. Polyphenol concentration was determined using the Prussian blue assay (Price & Butler, 1977; Goverde et al., 1999), and condensed tannin concentration was determined using the acid butanol assay (Porter et al., 1986; Goverde et al., 1999).

For cyanide analysis, leaf samples were frozen in liquid nitrogen immediately on removal from the plant and stored in a freezer at −80 °C until analysis. The samples were ground with liquid nitrogen in a mortar and pestle, and released cyanide was fixed in 0.5 ml of 1 N NaOH at 37 °C for 12 h. The cyanide concentration was estimated by the method of Lambert et al. (1975) and expressed in equivalent mg CN g−1 dry mass. A standard absorption curve at 725 nm was established with commercial KCN (Prolabo, Briare, France).

Statistical analysis

Plant growth data were analysed using a full factorial anova with CO2 treatment, plant genotype, and herbivory treatment as main factors. Shoot and root biomass were ln-transformed to improve normality. The residuals from a regression of shoot biomass vs. root biomass were used as an estimate of relative root/shoot dry matter partitioning (Goverde et al., 1999). Leaf allelochemicals (condensed tannins, total polyphenols, and cyanide) were analysed using a full factorial anova with CO2 treatment, plant genotype, herbivory treatment, and leaf age as main factors. Polyphenols and condensed tannins were ln-transformed to improve normality. Missing data due to sample damage during storage (32 of 91 samples for young leaves and 46 of 111 samples for old leaves) reduced the number of samples for the cyanide analysis. For each insect trait, a fully factorial model for CO2 treatment, plant genotype and maternal identity plus the factor sex of the individual was fitted.

Analyses were performed using SAS JMP III (SAS, 1995) using type III sums of squares. Stepwise reduction of all full models was used, removing highest–order interactions and always the least significant effect first (Crawley, 1993). Factors were only removed when P > 0.1. Model reduction stopped when no further interactions could be removed. No non-significant experimental main effects were removed.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Plant growth and allocation

The shoot and root biomass of L. corniculatus differed significantly between CO2 treatments. Shoot dry mass under elevated CO2 was 2.3 times greater than under ambient CO2 conditions, and root dry mass was 1.8 times greater under elevated than under ambient CO2 (Table 1, Fig. 1). Herbivory did not reduce the shoot biomass significantly, suggesting compensatory growth, but led to a marginal increase in root biomass (Table 1, Fig. 1).

Table 1.   Reduced model from full-factorial anova with carbon dioxide (CO2), plant genotype, and herbivory as main effects on ln-transformed shoot and root biomass and relative allocation to shoots, measured as the residuals from a regression of shoot on root biomass. * P < 0.05, *** P < 0.001.
SourceShoot biomassRoot biomassShoot/root
d.f.Fd.f.Fd.f.F
CO21102.89***134.33***125.12***
Plant genotype (P)355.11***321.50***37.17***
Herbivory (H)11.0713.7813.94*
CO2×P    33.13*
P×H    32.94*
Error125 126 117 
image

Figure 1.  Shoot and root dry mass and relative allocation to shoots, measured as the residuals from a regression of shoot on root biomass (mean ± SE, plant genotypes were pooled) of L. corniculatus plants grown under ambient (350 p.p.m.) and elevated (700 p.p.m.) carbon dioxide with (bsl00036) and without (bsl00000) herbivory.

Download figure to PowerPoint

The analysis of residuals from the regression of shoot on root biomass showed that shoot/root partitioning differed significantly between CO2 and herbivory treatments. Elevated CO2 concentrations led to decreased allocation to roots while herbivory led to increased allocation to roots (Table 1, Fig. 1). Furthermore, there was large genotypic variation in the response of plant dry matter allocation to CO2 enrichment and herbivory (significant plant genotype × CO2 and herbivore interaction respectively; Table 1).

Leaf chemistry

Plant genotypes differed significantly for all defence compounds because genotypes differing in carbon- and nitrogen-based defence compounds were chosen intentionally (Table 2). Plants grown under elevated CO2 conditions contained higher condensed tannin and polyphenol concentrations than plants grown under ambient CO2. Young leaves contained a higher concentration of these compounds than old leaves but there was no main effect of herbivory (Table 2, Fig. 2). For total polyphenol and condensed tannin, no difference between CO2 treatments was found for old leaves whereas for young leaves higher concentrations were found under elevated CO2 (significant CO2 × leaf age interaction; Table 2). The significant herbivory × leaf age interaction resulted from lower total polyphenol and condensed tannin concentrations under herbivory for old leaves but higher concentrations for young leaves.

Table 2.   Reduced model from full-factorial anova with carbon dioxide (CO2), plant genotype, herbivory, and leaf age as main effects on leaf concentrations of cyanide and ln-transformed total polyphenols and condensed tannins. * P < 0.05, ** P < 0.01, *** P < 0.001.
SourceTotal polyphenolsCondensed tanninsCyanide
d.f.Fd.f.Fd.f.F
CO2116.20***115.29***10.36
Plant genotype (P)322.85***322.19***37.88***
Herbivory (H)10.0010.01112.06***
Leaf age (L)178.11***177.05***171.14***
CO2×P    35.97***
CO2×L16.25*16.94**  
P×H    34.72**
H×L18.33**17.97**  
Error193 192 109 
image

Figure 2.  Total polyphenols, condensed tannins, and cyanide concentrations (mean ± SE, plant genotypes were pooled) in young and old leaves from L. corniculatus plants grown under ambient (350 p.p.m.) and elevated (700 p.p.m.) carbon dioxide with (bsl00036) and without (bsl00000) herbivory.

Download figure to PowerPoint

The cyanide concentration of leaves differed significantly between herbivory treatments and leaf age. No main effect of CO2 was detected but there was a significant interaction between plant genotype and CO2 (Table 2). Cyanide concentration was higher in young leaves than in old leaves and generally higher in plants without herbivory than in plants with herbivory (Fig. 2), however the latter was only true for three plant genotypes while the fourth genotype increased its cyanide concentration with herbivory (herbivory × plant genotype interaction; Table 2).

Growth response of Polyommatus icarus

Pupal fresh mass was significantly greater for caterpillars reared under elevated CO2 (102.32 ± 1.49 mg) than under ambient CO2 (97.14 ± 1.73 mg) and differed among plant genotypes, by maternal identity and by sex, with males weighing more than females at pupation (Table 3). Interestingly, for adult dry mass the effect of CO2 enrichment and plant genotype disappeared (Table 3) and females were heavier than males, indicating that male pupae contained more water than female pupae (see also Goverde et al., 1999).

Table 3.   Reduced model from full-factorial anova with carbon dioxide (CO2), plant genotype, and maternal identity as main effects plus the factor sex on pupal and adult mass and development time of caterpillars reared on food plants grown under ambient and elevated CO2. * P < 0.05, ** P < 0.01, *** P < 0.001.
SourcePupal weightAdult weightDevelopment time
d.f.Fd.f.Fd.f.F
CO2110.39**10.43127.35***
Plant genotype (P)311.32***31.4932.30
Maternal identity (M)35.61**32.89*32.77*
CO2×M    32.89*
Sex14.78*16.35*110.54**
Error77 71 74 

Total development time (time from first instar until adult eclosion) was shorter under elevated CO2 (29.6 ± 0.2 days) than under ambient CO2 (31.1 ± 0.3 days), and males developed faster than females (Table 3). Development time differed among maternal lines and marginally significantly among plant genotypes (Table 3). There was a significant interaction between maternal identity and CO2 treatment, with one maternal line revealing no reaction to CO2 enrichment.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

As predicted from the carbon/nutrient balance hypothesis, plants grown under elevated CO2 had higher biomass and more carbon-based defence compounds than plants grown under ambient CO2 conditions. The lack of a significant main effect of herbivory on shoot biomass indicated compensatory growth of these grazed plants, because caterpillars of P. icarus consume a significant mass of fresh foliar material each day (Goverde et al., 1999). Because no interaction between herbivory and CO2 treatment was found, this indicates no greater compensatory growth under elevated CO2, which corresponds to the results of some other studies (Fajer et al., 1991; Wilsey et al., 1994, 1997; Kruger et al., 1998), although in sugar maple regrowth was increased under elevated CO2 conditions (Kruger et al., 1998). Also, in contrast to the predictions, although plants grown under elevated CO2 increased their root biomass marginally, they decreased the relative allocation to root versus shoot growth (see also Briggs, 1991). Overall they contained comparable concentrations of cyanide in both CO2 treatments but this varied with plant genotype, two genotypes not changing, one increasing, and one decreasing. A final prediction of the carbon/ nutrient balance hypothesis was a greater difference in carbon-based defence molecules under the different CO2 and herbivory treatments, i.e. a significant interaction between herbivory and CO2 treatments. No such interaction was found, similar to findings on defence compounds for quaking aspen and sugar maple after simulated herbivory, although these species show such an interaction for protein concentration (Lindroth & Kinney, 1998). Although some of the presented results support the carbon/nutrient balance hypothesis, plants in the present study did not generally behave in accordance with this hypothesis, and this can be genotype specific. As discussed by Hamilton et al. (2001), the carbon/nutrient balance hypothesis impedes more than it contributes to understanding plant–consumer interactions. Alternatives are the protein competition model and the growth differentiation model (Herms & Mattson, 1992; Koricheva et al., 1998; Jones & Hartley, 1999; Bezemer et al., 2000).

Herbivory did not have the effects predicted by the carbon/nutrient balance hypothesis, perhaps because herbivory by these caterpillars affected plant architecture. Caterpillars fed preferentially on young leaves and apical buds, and most regrowth occurred via new lateral shoots, as in other forbs (e.g. Paige & Whitham, 1987). If the development of new meristems limited shoot regrowth, carbon may not have been a limiting factor, thus explaining the lack of a CO2 effect on compensatory growth. If this interpretation is correct, the carbon/nutrient balance hypothesis should not be applied without considering the effects of herbivores on meristems. Furthermore, as predicted by the protein competition model, phenylalanine might be the limiting resource so no effect of CO2 would be found (Jones & Hartley, 1999).

Plant defence variation by genotype will not be discussed because the genotypes were chosen to differ for these defence compounds (Goverde et al., 1999). Herbivory only affected cyanide concentrations, with grazed plants having less cyanide than control plants. If nitrogen is required for compensatory regrowth, less may be available for nitrogen-based defence (as suggested by Gleadow et al., 1998; Gleadow & Woodrow, 2000b). In addition, the fact that the different plant genotypes responded differently to herbivory demonstrates genetic variation for an induced response to herbivory for cyanide defence. It is not clear whether this variation is due to differences in the trade-off for nitrogen allocation to growth versus defence among genotypes or to different strategies of allocation per se.

Young leaves were better protected than old leaves for all defence compounds investigated, as is generally the case (e.g. Meyer & Montgomery, 1987; Williams et al., 1997; Gleadow & Woodrow, 2000b; for review, see Feeny, 1992). For carbon-based defence compounds, plants grown under elevated CO2 had much greater concentrations in young leaves, suggesting that the excess carbon was allocated specifically to defence in young leaves. Furthermore, in the absence of herbivory, there is little difference between young and old leaves. Plants exposed to herbivores show much greater concentrations of carbon-based defence compounds in young than in old leaves. Because young leaves are the principle targets of herbivores, this result implies induced response specifically for the tissues at highest risk (Raupp & Denno, 1983; Karban & Baldwin, 1997).

Caterpillars of P. icarus developed faster and attained higher pupal mass when reared on plants grown under elevated CO2 conditions, similar to previous findings (Goverde et al., 1999). Insect performance differed by plant genotype. The lowest pupal mass and longest developmental time were found on the genotype with the highest concentrations of both carbon- and nitrogen-based defence compounds. This was also the plant genotype with the lowest leaf biomass. Poor growth of caterpillars on this genotype could result from food limitation, because most young material was eaten, however the same plant genotype effect on insect growth was found by Goverde et al. (1999) when caterpillars were fed clipped leaf material ad libitum. This suggests that the genotype effect is qualitative, not quantitative. Furthermore, the concordance between the results from the two experiments suggests that, from the point of view of herbivores, the response of plants was similar whether plants were subjected to clipping or active herbivory. Although an induced response to herbivory was measured in the present study, the same plant response was apparently made to clipping. Thus, feeding experiments using clipped plant material can reasonably be compared with those using active herbivory (but see Rhoades, 1983).

Larvae of different parents differed in all characteristics, which is expected because these field-caught individuals differed for both genetic and environmental factors (Goverde et al., 1999, 2000). More interestingly, for larval developmental time, an important life-history trait, an interaction between maternal identity and CO2 was found. Larvae of one genotype did not accelerate their developmental time under elevated CO2. Carbon dioxide concentration can thus be considered a potential selective factor for this butterfly species if rapid development is advantageous (Gilbert & Singer, 1975) and there is genetic variation for the reaction of this trait to elevated CO2. It must be mentioned, however, that in the present study environmentally based maternal effects (Rossiter, 1991) were not controlled for, i.e. it is not known what the field-caught females were feeding on (as adults or as larvae) and how this could affect the outcome of the presented results.

In conclusion, in the present study little support was found for the carbon/nutrient balance hypothesis in L. corniculatus grown at different CO2 levels and with or without herbivory. It is therefore questionable whether this hypothesis should continue to be used to predict the effects of CO2 on plant–herbivore interactions (Hamilton et al., 2001); however evidence was found for genetic variation in the shoot versus root allocation patterns of plants in response to both herbivory and CO2. Similarly, plant genotypes differed in their production of cyanide in response to both herbivory and CO2. Caterpillar performance differed with CO2 treatment for the offspring of different mothers. Altogether this suggests that rising levels of CO2 might be a selective factor in both plant and herbivore populations if there is selection for shoot versus root allocation or cyanide defence for the plants, and developmental time for the caterpillars.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We are particularly grateful to O. Cudelou for technical support, and to P. Leadley, B. Baur and two anonymous referees for their valuable comments on this manuscript. The research was supported by the Priority Program Environment, Swiss National Science Foundation, grant no. 5001–044622/1 to A. Erhardt.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Awmack, C.S. & Harrington, R. (2000) Elevated CO2 affects the interactions between aphid pests and host plant flowering. Agricultural and Forest Entomology, 2, 5761.
  • Ayres, M.P. (1993) Plant defense, herbivory, and climate change. Biotic Interactions and Global Change (ed. by P. M.Kareiva, J. G.Kingsolver and R. B.Huey), pp. 7594. Sinauer Associates, Sunderland, Massachusetts.
  • Bezemer, T.M. & Jones, T.H. (1998) Plant–insect herbivore interactions in elevated atmospheric CO2: quantitative analyses and guild effects. Oikos, 82, 212222.
  • Bezemer, T.M., Jones, T.H. & Newington, J.E. (2000) Effects of carbon dioxide and nitrogen fertilization on phenolic content in Poa annua L. Biochemical Systematics and Ecology, 28, 839846.
  • Briggs, M.A. (1990) Chemical defense production in Lotus corniculatus L. I. The effect of nitrogen source on growth, reproduction and defense. Oecologia, 83, 2731.
  • Briggs, M.A. (1991) Influence of herbivory and nutrient availability on biomass, reproduction and chemical defences in Lotus corniculatus. Functional Ecology, 5, 780786.
  • Briggs, M.A. & Schultz, J.C. (1990) Chemical defense production in Lotus corniculatus L. II. Trade-offs among growth, reproduction and defense. Oecologia, 83, 3237.
  • Bryant, J.P., Chapin, F.S. & Klein, D.R. (1983) Carbon/nutrient balance of boreal plants in relation to vertebrate herbivory. Oikos, 40, 357368.
  • Bryant, J.P., Heitkonig, I., Kuropat, P. & Owen-Smith, N. (1991) Effect of severe defoliation on the long term resistance to insect attack and on the leaf chemistry in six woody species of the southern African savanna. American Naturalist, 137, 5063.DOI: 10.1086/285145
  • Bultman, T.L. & Conard, N.J. (1998) Effects of endophytic fungus, nutrient level, and plant damage on performance of fall armyworm (Lepidoptera: Noctuidae). Environmental Entomology, 27, 631635.
  • Coley, P.D., Bryant, J.P. & Chapin, F.S. (1985) Resource availability and plant antiherbivore defense. Science, 230, 895899.
  • Coviella, C.E. & Trumble, J.T. (1999) Effects of elevated atmospheric carbon dioxide on insect–plant interactions. Conservation Biology, 13, 700712.
  • Crawley, M.J. (1993) GLIM for Ecologists. Blackwell Science, Oxford.
  • Fajer, E.D., Bowers, M.D. & Bazzaz, F.A. (1989) The effects of enriched carbon dioxide atmospheres on plant–insect herbivore interactions. Science, 243, 11981200.
  • Fajer, E.D., Bowers, M.D. & Bazzaz, F.A. (1991) Performance and allocation patterns of the perennial herb Plantago lanceolata, in response to simulated herbivory and elevated CO2 environments. Oecologia, 87, 3742.
  • Fajer, E.D., Bowers, M.D. & Bazzaz, F.A. (1992) The effect of nutrients and enriched CO2 environments on production of carbon-based allelochemicals in Plantago: a test of the carbon/nutrient balance hypothesis. American Naturalist, 140, 707723.
  • Feeny, P. (1992) The evolution of chemical ecology: contributions from the study of herbivorous insects. Herbivores. Their Interaction with Secondary Plant Metabolites (ed. by G. A.Rosenthal and M. R.Berenbaum), pp. 144. Academic Press, San Diego, California.
  • Gilbert, L.E. & Singer, M.C. (1975) Butterfly ecology. Annual Review of Ecology and Systematics, 6, 365397.
  • Gleadow, R.M., Foley, W.J. & Woodrow, I.E. (1998) Enhanced CO2 alters the relationship between photosynthesis and defence in cyanogenic Eucalyptus cladocalyx F. Muell. Plant, Cell and Environment, 21, 1222.
  • Gleadow, R.M. & Woodrow, I.E. (2000a) Polymorphism in cyanogenic glycoside content and cyanogenic beta-glucosidase activity in natural populations of Eucalyptus cladocalyx. Australian Journal of Plant Physiology, 27, 693699.
  • Gleadow, R.M. & Woodrow, I.E. (2000b) Temporal and spatial variation in cyanogenic glycosides in Eucalyptus cladocalyx. Tree Physiology, 20, 591598.
  • Goverde, M., Bazin, A., Shykoff, J.A. & Erhardt, A. (1999) Influence of leaf chemistry of Lotus corniculatus (Fabaceae) on larval development of Polyommatus icarus (Lepidoptera, Lycaenidae): effects of elevated CO2 and plant genotype. Functional Ecology, 13, 801810.DOI: 10.1046/j.1365-2435.1999.00372.x
  • Goverde, M., Erhardt, A. & Niklaus, P.A. (2002) In situ development of a satyrid butterfly on calcareous grassland exposed to elevated carbon dioxide. Ecology, in press.
  • Goverde, M., Van Der Heijden, M.G.A., Wiemken, A., Sanders, I.R. & Erhardt, A. (2000) Arbuscular mycorrhizal fungi influence life history traits of a lepidopteran herbivore. Oecologia, 125, 362369.
  • Hamilton, J.G., Zangerl, A.R., DeLucia, E.H. & Berenbaum, M.R. (2001) The carbon-nutrient balance hypothesis: its rise and fall. Ecology Letters, 4, 8695.DOI: 10.1046/j.1461-0248.2001.00192.x
  • Havill, N.P. & Raffa, K.F. (1999) Effects of elicitation treatment and genotypic variation on induced resistance in Populus: impacts on gypsy moth (Lepidoptera: Lymantriidae) development and feeding behaviour. Oecologia, 120, 295303.
  • Herms, D.A. & Mattson, W.J. (1992) The dilemma of plants: to grow or to defend. Quarterly Review of Biology, 67, 283335.
  • Houghton, J.T., Meira Filho, L.G., Callander, B.A., Harris, N., Kattenberg, A. & Maskell, K. (1996) Climate Change 1995. The Science of Climate Change. Cambridge University Press, Cambridge.
  • Jones, C.G. & Hartley, S.E. (1999) A protein competition model of phenolic allocation. Oikos, 86, 2744.
  • Jones, D.A. (1988) Cyanogenesis in animal–plant interactions. Cyanide Compounds in Biology (ed. by D.Evered and S.Harnett), pp. 151165. J. Wiley & Sons, Chichester, U.K.
  • Karban, R. (1993) Induced resistance and plant density of a native shrub, Gossypium thurberi, affect its herbivores. Ecology, 74, 18.
  • Karban, R. & Baldwin, J.T. (1997) Induced Responses to Herbivory. Chicago Press, Chicago, Illinois.
  • Kinney, K.K., Lindroth, R.L., Jung, S.M. & Nordheim, E.V. (1997) Effects of CO2 and NO3– availability on deciduous trees: phytochemistry and insect performance. Ecology, 78, 215230.
  • Koricheva, J., Larsson, S., Haukioja, E. & Keinämen, M. (1998) Regulation of woody plant secondary metabolism by resource availability: hypothesis testing by means of meta-analysis. Oikos, 83, 212226.
  • Kruger, E.L., Volin, J.C. & Lindroth, R.L. (1998) Influences of atmospheric CO2 enrichment on the responses of sugar maple and trembling aspen to defoliation. New Phytologist, 140, 8594.
  • Lambert, J.L., Ramasamy, J. & Paukstelis, J.V. (1975) Stable reagents for the colorimetric determination of cyanide by the modified Konig reactions. Analytical Chemistry, 47, 916918.
  • Lincoln, D.E., Fajer, E.D. & Johnson, H.J. (1993) Plant–insect herbivore interactions in elevated CO2 environments. Trends in Ecology and Evolution, 8, 6468.
  • Lindroth, R.L., Arteel, G.E. & Kinney, K.K. (1995) Responses of three saturniid species to paper birch grown under enriched CO2 atmospheres. Functional Ecology, 9, 306311.
  • Lindroth, R.L. & Kinney, K.K. (1998) Consequences of enriched atmospheric CO2 and defoliation for foliar chemistry and gypsy moth performance. Journal of Chemical Ecology, 24, 16771695.
  • Lovelock, C.E., Posada, J. & Winter, K. (1999) Effects of elevated CO2 and defoliation on compensatory growth and photosynthesis of seedlings in a tropical tree, Copaifera aromatica. Biotropica, 31, 279287.
  • Mattson, W.J. (1980) Herbivory in relation to plant nitrogen content. Annual Review of Ecology and Systematics, 11, 119161.
  • Meyer, G.A. & Montgomery, M.E. (1987) Relationships between leaf age and the food quality of cottonwood foliage for the gypsy moth, Lymantria dispar. Oecologia, 72, 527532.
  • Paige, K.N. & Whitham, T.G. (1987) Overcompensation in response to mammalian herbivory: the advantage of being eaten. American Naturalist, 129, 407416.
  • Penuelas, J. & Estiarte, M. (1998) Can elevated CO2 affect secondary metabolism and ecosystem function? Trends in Ecology and Evolution, 13, 2024.
  • Porter, L.J., Hrstich, L.N. & Chan, B.G. (1986) The conversion of procyanidins and prodelphinidins to cyanidin and delphinidin. Phytochemistry, 25, 223230.DOI: 10.1016/S0031-9422(00)94533-3
  • Price, M.L. & Butler, L.G. (1977) Rapid visual estimations and spectrophotometric determination of tannin content of sorghum grain. Journal of Agriculture and Food Chemistry, 25, 12691273.
  • Raupp, M.J. & Denno, R.F. (1983) Leaf age as a predictor of herbivore distribution and abundance. Variable Plants and Herbivores in Natural and Managed Systems (ed. by R. F.Denno and M. S.McClure), pp. 91124. Academic Press, New York.
  • Rhoades, D.F. (1983) Herbivore population dynamics and plant chemistry. Variable Plants and Herbivores in Natural and Managed Systems (ed. by R. F.Denno and M. S.McClure), pp. 155220. Academic Press, New York.
  • Rieske, L.K. & Raffa, K.F. (1998) Interactions among insect herbivore guilds: influence of thrips bud injury on foliar chemistry and suitability to gypsy moths. Journal of Chemical Ecology, 24, 501523.
  • Rossiter, M.C. (1991) Environmentally-based maternal effects: a hidden force in insect population dynamics? Oecologia, 87, 288294.
  • Ryle, G.J.A. & Powell, C.E. (1992) The influence of elevated carbon dioxide and temperature on biomass production of continuously defoliated white clover. Plant, Cell and Environment, 15, 593599.
  • SAS (1995) JMP®User's Guide, Version 3.1. SAS Institute, Cary, North Carolina.
  • Tuomi, J., Niemelä, P., Chapin, F.S., Bryant, J.P. & Siren, S. (1988) Defensive responses of trees in relation to their carbon/nutrient balance. Mechanisms of Woody Plant Defences Against Insects (ed. by W. J.Mattson, J.Levieux and C.Bernard-Dagan), pp. 5772. Springer Verlag, New York.
  • Tuomi, J., Niemelä, P. & Siren, S. (1990) The Panglossian paradigm and delayed inducible accumulation of foliar phenolics in mountain birch. Oikos, 59, 399410.
  • Watt, A.D., Whittaker, J.B., Docherty, M., Brooks, G., Lindsay, E. & Salt, D.T. (1995) The impact of elevated atmospheric CO2 on insect herbivores. Insects in a Changing Environment (ed. by R.Harrington and N. E.Stork), pp. 197217. Academic Press, London.
  • Whittaker, J.B. (1999) Impacts and responses at population level of herbivorous insects to elevated CO2. European Journal of Entomology, 96, 149156.
  • Williams, R.S., Lincoln, D.E. & Norby, R.J. (1998) Leaf age effects of elevated CO2-grown white oak leaves on spring-feeding lepidopterans. Global Change Biology, 4, 235246.
  • Williams, R.S., Thomas, R.B., Strain, B.R. & Lincoln, D.E. (1997) Effects of elevated CO2, soil nutrient levels, and foliage age on the performance of two generations of Neodiprion lecontei (Hymenoptera: Diprionidae) feeding on loblolly pine. Environmental Entomology, 26, 13121322.
  • Wilsey, B.J., Coleman, J.S. & McNaughton, S.J. (1997) Effects of elevated CO2 and defoliation on grasses: a comparative ecosystem approach. Ecological Applications, 7, 844853.
  • Wilsey, B.J., McNaughton, S.J. & Coleman, J.S. (1994) Will increases in atmospheric CO2 affect regrowth following grazing in C4 grasses from tropical grasslands? A test with Sporobolus kentrophyllus. Oecologia, 99, 141144.

Accepted 20 September 2001