SEARCH

SEARCH BY CITATION

Keywords:

  • Herbivory;
  • host plant diversity;
  • host plant nitrogen;
  • larval survival;
  • mixed diet;
  • Stenodemini

Abstract

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

Abstract. 1. In a laboratory experiment, the influence of host plant diversity and food quality, in terms of nitrogen content, on the larval survival of two oligophagous bug species (Heteroptera, Miridae: Leptopterna dolobrata L., Notostira erratica L.) was investigated. Both species are strictly phytophagous and capable of feeding on a wide range of grass species. Moreover, they typically change their host plants during ontogenesis; it has been suggested that this behaviour is a response to the changing protein content of the hosts.

2. To investigate the importance of host plant diversity for these insects, the development of insects reared on grass monocultures was compared with that on mixtures of four grass species. In addition, the host grasses were grown under two nitrogen regimes to test whether nitrogen content is the key factor determining host plant switching.

3. Both species had a significantly higher survival rate when feeding on several host plants but only L. dolobrata showed a significant response to food nitrogen content. Furthermore, there was no correlation between the nitrogen content of the host plants and the survival rate of N. erratica larvae.

4. The study suggests that at least some Stenodemini need a variety of host plants during larval development but that the level of host plant nitrogen is not the main factor responsible for the observed diversity effect.


Introduction

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

In recent years, there has been considerable interest in how species diversity influences ecosystem function. Various experimental designs have been used to investigate this question, including field plots with different numbers of plant species (Tilman et al., 1996; Hector et al., 1999) and laboratory micro- and meso-cosms in which diversity has been manipulated at a variety of trophic levels (Naeem et al., 1994). Although the interpretation of such experiments remains controversial (Huston, 1997; Wardle, 1999), a common finding has been that a high level of species diversity is associated with a higher level of ecosystem performance, as reflected in parameters such as biomass production and ecosystem respiration. There has been little research, however, on the relationship between plant species diversity and the performance of insect herbivores.

Herbivory presents major difficulties to insects, partly because the chemical composition of their tissues is very different from that of plants. In particular, the protein content of insects is much higher than that of plants and the proportions of the various amino acids differ between animal and plant proteins (Southwood, 1973; Mattson, 1980; White, 1993). It is well known that the nitrogen content of plant food can be a crucial factor for the development and reproduction of herbivores (Strong et al., 1984). In addition, the allocation of nutrients to structures like stems, fruits, and flowers changes considerably both seasonally and during the course of plant development. Thus, the protein content of plants is not only generally low but also varies during the season. Some phytophagous insects apparently respond to changes in nutrient availability by switching from a plant with a low protein level to a plant with a higher protein concentration (McNeill & Southwood, 1978). This could be one reason why some polyphagous insects perform best when feeding on a mixture of plants rather than a single plant species (Bernays et al., 1992, 1994; Modder & Tamu, 1996; Hägele & Rowell-Rahier, 1999), though not all polyphagous species react similarly (Bernays & Minkenberg, 1997). Few studies have investigated the effects of a mixed diet on oligophagous insects (Ballabeni & Rahier, 2000).

The effect of plant species diversity on the performance of two oligophagous species of Stenodemini (Heteroptera: Miridae), Leptopterna dolobrata L. and Notostira erratica L., was investigated. The Stenodemini are phytophagous insects that feed exclusively on grasses and change their host plants during ontogenesis. They are normally found on a restricted set of grass species, though it is known that they can feed on a much wider range. It has been suggested that the host switching is a response to the changing protein content of their hosts (Braune, 1971; McNeill, 1971, 1973; Gibson, 1976, 1980). The larval survival of insects reared on grass monocultures was compared with that of insects reared on mixtures of four species. To test whether total nitrogen content is the key factor responsible for host plant switching, the plants were cultivated under two nitrogen regimes. It is known that total nitrogen content of grasses tends to increase with nitrogen fertilisation, though the reaction of individual grass species is variable (Meister & Lehmann, 1990).

Methods

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

Experimental design

The aim of the experiment was to investigate the effect of plant species diversity on larval survival. The important variable was therefore the number of species per se rather than which species were present in the mixture; replicates were therefore represented by several different combinations of the same number of species (Tilman et al., 1996). Monocultures were compared with mixtures of four species. The species in both the monocultures and the mixtures were chosen randomly out of a species pool comprising eight common grass species that are known to be host plants of the bugs (Table 1). Total nitrogen content was included in the experimental design as a second factor in order to separate the nitrogen effect from the diversity effect (Huston, 1997). The low nitrogen level was selected to correspond approximately to the typical nutrient input in an extensively managed meadow (i.e. no fertiliser input), and the high level to one of an intensively managed meadow (40–60 m3 liquid manure per ha and year). A fully factorial design was used with two factors, plant diversity and nitrogen content, both of which had two levels. Thus insects were offered either one or four plant species, which had been raised on either high or low nitrogen. There were five replicates for each treatment (Table 2). To compare the single monocultures with the mixtures, and to exclude the sampling effect (Wardle, 1999), the three monocultures that were not in the sampling design (P. trivialis, F. ovina, F. pratensis) were also examined. As there were not enough larvae of L. dolobrata, all monocultures were only tested with N. erratica. The insects were reared on plants grown with high nitrogen (Table 2).

Table 1.  The grass species selected for the experiment and the randomly chosen monocultures and mixtures in the experimental design. Five monocultures and five mixtures were chosen randomly from a species pool comprising eight grass species. All species are quoted in the literature as host plants of Notostira erratica and Leptopterna dolobrata .
  Mixtures 
SpeciesMonocultures12345References
Arrhenaherum elatius××××× Kullenberg (1944) ; Gibson (1976 )
Dactylis glomerata××   ×Osborn (1918) ; Kullenberg (1944) ; Southwood & Leston (1959); Braune (1971); Gibson (1976)
Festuca ovina  × × Osborn (1918) ; Kullenberg (1944) ; Gibson (1976 )
Festuca pratensis × ×  Osborn (1918) ; Kullenberg (1944) ; Gibson (1976 )
Holcus lanatus×  ×× Kullenberg (1944) ; Braune (1971 ); Gibson (1976 )
Lolium perenne×××× ×Kullenberg (1944)
Phleum pratense×   ××Kullenberg (1944) ; Southwood & Leston (1959)
Poa trivialis  ×  ×Kullenberg (1944)
Table 2.  Experimental design with the number of replicates per treatment and bug species.
TreatmentHigh N levelLow N levelBug species
Monoculture85Notostira erratica
55Leptopterna dolobrata
Mixture55Notostira erratica
55Leptopterna dolobrata

Cultivation and chemical analysis of the food plants

The food grasses were cultivated from the end of March to the end of August. The grass seeds were of local provenance and provided by the Swiss Federal Research Station of Agroecology and Agriculture. The cultivation method is a standardised procedure used at the research station to study the effects of different nutrient inputs on the growth and development of various grass and leguminous species (U. Walther, pers. comm.). The plants were grown out of doors under a mobile roof. They were uncovered in dry weather but remained covered when it was raining so that water and nutrient input could be controlled, while temperature, humidity, and light regime corresponded closely to natural conditions. Each species was cultivated separately in double-walled pots of 23.5 cm diameter. The pots were filled with 9 kg of loamy soil that was fertilised with 6 g K2HPO4 before sowing. The nitrogen fertiliser was prepared as a solution (2.5 g l−1 of CaN2O6 × 4H2O), and 100 ml was added to a pot at any one time. The pots were watered regularly and any surplus water that drained through the pots was collected in separate containers, from where it was refilled into the pots; in this way no nutrients were lost. The grasses grown at a low nitrogen level were fertilised only once, 10 weeks after germination. The pots of the high nitrogen treatment were fertilised monthly and thus received five times as much nitrogen over the whole experimental period.

The food plants for the insects and the plant material for the chemical analysis came from the same batches of material and the same plant parts that were given as food to the insects were analysed. The plant material was analysed five times from the beginning of May to mid August For the analysis, leaves and stems of each grass species were collected every 2 or 3 weeks (Fig. 1). The total nitrogen content of the plant material was determined by a modification of the Kjeldahl method (Novozamsky et al., 1983; Houba et al., 1989).

image

Figure 1. The total nitrogen content of the two treatments differed significantly ( anova , F1,78  = 24.89, P  < 0.001) and varied during the experimental period. The grasses with high nitrogen input (upper graph) were fertilised regularly; the grasses with low nitrogen input (lower graph) were fertilised once in May (see text for specification). The grey boxes show approximately the period during which the two bug species developed. The dark grey box indicates the development time of Leptopterna dolobrata (1–26 June), the light grey box the development time of Notostira erratica (22 July−25 August). The protein level was high but varied strongly in late spring when L. dolobrata developed, while it was low and varied less in summer, when N. erratica developed.

Download figure to PowerPoint

Rearing the insects

The insect rearing cages were 25 cm cubes made from acrylic Plexiglas®. They had a circular opening in the front fitted with a gauze tube for handling the insects. Each cage was equipped with a small pot to hold the food plants, and a water dispenser that was filled daily. Food plants for both bug species came from the same batches of material over the whole experimental period. Freshly cut tillers served as food and were provided every second day. The cages were checked daily and food plants that wilted were replaced immediately. Only tillers were fed to both bug species, although under natural conditions older larvae of L. dolobrata (instars 3–5) also feed on flowers and seeds. All animals received eight tillers at any one time. For the mixtures, two tillers of each grass species were combined. The specimens were reared under a constant LD 16:8 h photoperiod at a constant 20 °C. The humidity in the climate chamber varied between 60 and 90%. The cages were sprayed regularly with water to raise the humidity inside the boxes and to avoid negative effects of low humidity on larval survival. To minimise positional bias within the climate chamber, the positions of the cages were changed regularly. Every day, any dead insects were removed and the number and stage of the larvae were determined. The adult insects were killed the day after the final moult.

The experimental material for L. dolobrata comprised first-instar larvae, which were collected in the field at the beginning of June from three populations in Safien, Canton Graubünden, Switzerland (9°19′E, 46°41′N). Eight larvae were placed in each cage immediately after collection (Fig. 1). The state of development of the larvae was recorded daily. The duration from moult to moult and the total duration (20–22 days) corresponded closely with the values given by McNeill (1971). Larvae of N. erratica were reared from adult females that were collected at the beginning of July in the Schaffhauser Randen, Canton Schaffhausen, Switzerland (8°36′E, 47°45′N). These females were kept in the climate chamber and provided with freshly cut grass material. In the first few days, several females were observed laying eggs on the upper part of the stems. The stems were replaced every second day and those with eggs were placed on humid tissues and kept in transparent plastic boxes in the climate chamber. The first larvae hatched after 12–14 days. Most larvae hatched within 3 days and were placed immediately in the cages. Thus, by the end of July and within 3 days the experiment was started. The first males became adult after 22 days and the first females after 25 days; all insects had reached adulthood by 35 days (Fig. 1).

Statistical analysis

One-way anova was used to test how nitrogen fertilisation affected the total nitrogen content of the grasses during the season. Two-way anova was used to analyse the influence of total nitrogen content and host plant diversity on larval survival of N. erratica and L. dolobrata. As a measure of larval survival, the proportion of the larvae that emerged as adults was calculated. All values were arcsin transformed for the analysis. Two models for the N. erratica data were tested. The first model comprised all monocultures tested, while the second model included just the five monocultures used in the experimental design (Table 2). Two models were also tested for the data of L. dolobrata. Only individuals reaching adulthood were included in the first model. Because mortality during the final moult was high and the total number of insects was rather small, a model that comprised the individuals that died during the final moult was also tested.

Results

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

Food quality

Nitrogen concentrations were significantly higher in grasses receiving the high nitrogen treatment than in grasses grown at low nitrogen level (anova, F1,78 = 24.89, P < 0.001), but the differences in nitrogen concentration between plant species were greater among plants grown with low nitrogen (Fig. 1). For example, the concentration of nitrogen in late May ranged from 30.4 g N kg−1 dry weight (A. elatius) to 38.8 g N kg−1 dry weight (P. trivialis) under the high nitrogen treatment, but from 13.0 g N kg−1 dry weight (L. perenne) to 40.4 g N kg−1 dry weight (D. glomerata) under the low nitrogen treatment. There were also differences between the two nitrogen treatments in the temporal variability in nitrogen content. Under both treatments, the nitrogen content dropped in all species between June and July, but this effect was much more pronounced under the high nitrogen treatment. Because the two bug species were not raised at the same time of year, they experienced differing ranges of nitrogen concentrations during the rearing experiments. During May and June, when L. dolobrata was raised, the nitrogen content was on average higher but more variable than in July and August when N. erratica was studied (Fig. 1).

Influence of host plant diversity and food quality on larval survival of L. dolobrata

No larvae of L. dolobrata survived on monocultures on either the low or high nitrogen plants (Fig. 2). On mixtures, larval survival was 12.5–75% on high nitrogen plants and 0–25% on low nitrogen plants. There were also differences in the timing of larval mortality on monocultures compared with mixtures (Fig. 3). Whereas the majority of insects reared on monocultures died before they had reached the fifth larval stage (75%), mortality of insects reared on mixtures was distributed more equally between the larval stages.

image

Figure 2. (a) Interaction plot of the two-way ANOVA of the effect of nitrogen content and plant diversity on larval survival of Leptopterna dolobrata. For this analysis, the insects that died at the final moult were included (Model I). The nitrogen level represents the two nitrogen regimes: high and low fertiliser input. Mean values of the two treatments are shown. (b) The larval survival of L. dolobrata increased with the species number of host plants. Larval survival of insects reared on monocultures was compared with that of insects reared on mixtures of four species. The dots indicate the replicates of different species combinations at each level of species richness. The size of the dots represents the number of cases.

Download figure to PowerPoint

image

Figure 3. The mortality of Leptopterna dolobrata at different larval stages. More larvae of stages 2–4 died when reared on monocultures than when reared on mixtures of four species.

Download figure to PowerPoint

The anova confirmed that both host plant diversity (Model I: F1,16 = 22.86, P < 0.001; Model II: F1,16 = 7.8, P < 0.05) and nitrogen treatment (Model I: F1,16 = 13.91, P < 0.01; Model II: F1,16 = 3.91, P = NS) affected larval survival significantly. There was also a significant interaction between the two factors (Model I: F1,16 = 13.91, P < 0.01; Model II: F1,16 = 3.91, P = NS; Fig. 2a).

Influence of host plant diversity and food quality on larval survival of N. erratica

Host plant diversity increased the survival rate of N. erratica very greatly (F1,16 = 22.66, P < 0.001) while nitrogen treatment had no effect (F1,16 = 0.21, P = NS). There was no interaction between nitrogen content and diversity (F1,16 = 0.27, P = NS; Fig. 4a). The same effects were found when including all monocultures in the model (diversity: F1,16 = 16.28, P < 0.001; nitrogen level: F1,16 = 0.55, P = NS; interaction term: F1,16 = 0.63, P = NS). Survival on monocultures was 0–88%, while on mixtures it was 50–100% (Fig. 4b). On four monocultures, survival was actually higher than on mixtures (F. ovina 88%, F. pratensis 82%, L. perenne 67 and 71%). It is interesting that the mixtures with the lowest survival rates all included one or more of these grasses. For example, although F. ovina was the monoculture supporting the highest survival rate, it was present in the two mixtures with the lowest survival rates (50 and 67%). In both nitrogen treatments, larval mortality occurred mainly during the first or second larval stages (Fig. 5).

image

Figure 4. (a) Interaction plot of the two-way anova of the effect of nitrogen content and plant diversity on larval survival of Notostira erratica. For this analysis, the fully factorial design was chosen with five monocultures and five mixtures for both nitrogen levels. The nitrogen level represents the two nitrogen regimes high and low fertiliser input. Mean values of the two treatments are shown. (b) The larval survival of N. erratica increased with the species number of the host plants. Larval survival of insects reared on monocultures was compared with that of insects reared on mixtures of four species. All monocultures tested in the experiment were included in this analysis, i.e. five monocultures of low nitrogen plants and eight of high nitrogen plants. The dots indicate the replicates of different species combinations at each level of species richness. The size of the dots represents the number of cases.

Download figure to PowerPoint

image

Figure 5. The mortality of Notostira erratica during larval development. Many more young larvae of the first and second stage died when reared on monocultures than when reared on mixtures of four species. Larvae reared on mixtures generally had very low mortality. For this analysis, the fully factorial design was chosen with five monocultures and five mixtures.

Download figure to PowerPoint

The mean nitrogen content of plant material was calculated as the average of the nitrogen content at the beginning of the experiment (27 July) and after 2 weeks (12 August). There was no significant correlation between larval survival of N. erratica and the mean nitrogen content of the food plants (t6 = −0.88, P = NS).

Discussion

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

Host plant diversity

Both Stenodemini species had a significantly higher survival rate on average when feeding on four host plants than when feeding on only one host plant; however, only in the case of L. dolobrata is there evidence that survival was affected by the nitrogen content of the plant material (Figs 2 and 4). These results suggest that neither the absolute level nor variation of host plant nitrogen is the main factor responsible for host plant switching, as has been hypothesised (McNeill, 1971, 1973; Gibson, 1976, 1980). Other factors probably account for the positive effect of host plant diversity on the survival of the two bug species. Previous studies have indicated that insects respond to several components of the diet, including the chemical form of nitrogen, the balance of amino acids, the water content of the host, and the content of minerals and trace elements (McNeill & Southwood, 1978; Strong et al., 1984; Bernays et al., 1992, 1994; Hägele & Rowell-Rahier, 1999). Most plant-feeding Heteroptera harbour symbiotic bacteria in the gut that synthesise sterols and amino acids. Like all Miridae, however, the Stenodemini have no symbiotic bacteria and may therefore be especially dependent on the nutrient balance of their plant food (Dolling, 1991). Secondary compounds and mechanical or chemical plant defences may also play an important role (McNeill & Southwood, 1978; Strong et al., 1984; Bernays et al., 1992, 1994; Bernays & Minkenberg, 1997; Hägele & Rowell-Rahier, 1999). Hoffman and McEvoy (1986) showed that the meadow spittlebug Philaenus spumarius was limited to feeding on host plants by mechanical plant defences such as trichomes and tissue hardness.

It cannot be determined from this experiment which factors caused the positive effect of a mixed diet. There is some indication that physical or chemical plant defences may have prevented the young larvae from feeding on some plant species. Notostira erratica larvae reared on monocultures showed the highest mortality in the first larval stage, while mortality was much lower later in development (Fig. 5). Mortality of insects reared on monocultures of A. elatius (100%), H. lanatus (100%), and D. glomerata (77.5%) was extremely high, although these plants are usual hosts of older larvae and adults. First-instar larvae died very soon after they were put into the cages, indicating that they could not feed on these grasses; however larvae reared on mixtures could avoid the negative effects of the plants by choosing among various species and therefore had a better chance of survival. These findings are supported by the work of Gibson (1976), who observed that certain grass species are avoided most strongly by very young Stenodemini larvae. Hoffman and McEvoy (1986) found that the young larvae of meadow spittlebugs in particular were limited by mechanical barriers and that the range of feeding sites increased with instar.

A problem inherent in all diversity experiments is the so-called sampling effect; in this experiment, a positive effect of mixtures could result simply because there is a higher probability that mixtures of several species contain a good quality plant, in terms of high nitrogen content, rather than monocultures (Huston, 1997); however the results indicate that the diversity effect cannot be explained entirely in this way. No larvae of L. dolobrata survived on monocultures, while most of the larvae reared on mixtures of high nitrogen plants did reach the adult stage (Fig. 2). The effect is less clear for N. erratica, though in most cases survival on mixtures was greater than on any of the monocultures (Fig. 4); however the values of four mixtures lie below those of the best monocultures even though grass species of the best four monocultures are also included in these mixtures. This suggests that larval survival on mixtures is not determined solely by the best plant species in the mixture but depends on the variety of plant species present.

Nitrogen

Larval survival of N. erratica was not affected by either the nitrogen treatment or the nitrogen content of the host plants (Fig. 4), suggesting that it can cope equally well with a wide range of nitrogen levels (Gibson, 1976). The food plants of N. erratica had much lower nitrogen contents than those of L. dolobrata, mainly because the nitrogen content of the grasses drops towards mid-summer and remains low until autumn (Fig. 1). These trends occur also in natural populations of grass (Gibson, 1980). Notostira erratica produces two generations per year and the second generation, which emerges in late summer, appears to be able to cope with very low levels of dietary nitrogen. Indeed, Gibson (1980) showed that N. erratica does not exploit the nitrogen available in a sward fully and apparently does not select within a sward for plants with above-average nitrogen content.

In contrast, L. dolobrata does seem to need both host plant diversity and plant material of good quality (Fig. 2), and several studies have shown that it has a high nitrogen demand during its development. Gibson (1980) demonstrated that L. dolobrata exploits the available nitrogen in a sward fully by feeding on hosts with above-average nitrogen content. Furthermore, it changes during the season from leaf to seed feeding when plant nutrients are allocated to the development of stems and seeds and the protein content of leaves drops sharply (McNeill, 1971, 1973). In the experiment, only tillers were fed, which probably increased the nitrogen effect found. The majority of the larvae reared on monocultures died between the second and fourth larval stage while, when reared on mixtures, larval mortality was distributed more evenly over the development time (Fig. 3). This indicates that the second to fourth larval instars are sensitive stages, during which food quality and a variety of host plants play an important role.

To summarise, the experiments show that at least some Stenodemini, like many polyphagous insects, are dependent on a variety of host plants for their larval development and survival but that the level of host plant nitrogen is not the main factor responsible for the observed diversity effect.

Acknowledgements

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

We thank Josef Lehmann, Andreas Rüegger and Hansueli Briner for the grass seeds, Josef Lehmann, Ueli Walther, Ernst Brack and Robert Richli for their assistance and tips on cultivating the food grasses and Stefan Bosshard for his help in rearing the bugs in the climate chamber. We are grateful to Franz Schubiger and the technical staff of the Swiss Federal Research Station for Agroecology and Agriculture for the chemical analysis of the plant material, to Josef Lehmann and Walter Dietl for their comments on the reaction of grasses to fertilisation and to Bernhard Merz, Sibylle Studer and Erhard Meister for their critical reading of the manuscript. Further we thank two anonymous reviewers for their useful comments on an earlier version of the manuscript. Finally, we thank Bernhard Schmid for his assistance with the experimental design. This work was funded by the Swiss National Science Foundation (5001-044634/1) and the Canton Schaffhausen (Kantonales Planungs- und Naturschutzamt).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Ballabeni, P. & Rahier, M. (2000) A quantitative genetic analysis of leaf beetle larval performance on two natural hosts: including a mixed diet. Journal of Evolutionary Biology, 13, 98106.
  • Bernays, E.A., Bright, K.L., Gonzalez, N. & Angel, J. (1994) Dietery mixing in a generalist herbivore: tests of two hypotheses. Ecology, 75, 19972006.
  • Bernays, E.A., Bright, K., Howard, J.J., Raubenheimer, D. & Champagne, D. (1992) Variety is the spice of life: frequent switching between foods in the polyphagous grasshopper Taeniopoda eques Burmeister (Orthoptera: Acrididae). Animal Behaviour, 44, 721731.
  • Bernays, E.A. & Minkenberg, O.P.J.M. (1997) Insect herbivores: different reasons for being a generalist. Ecology, 78, 11571169.
  • Braune, H.-J. (1971) Der Einfluss der Temperatur auf Eidiapause und Entwicklung von Weichwanzen (Heteroptera, Miridae). Oecologia, 8, 223266.
  • Dolling, W.R. (1991) The Hemiptera. Oxford University Press, Oxford.
  • Gibson, C.W.D. (1976) The importance of foodplants for the distribution and abundance of some Stenodemini (Heteroptera: Miridae) of limestone grasslands. Oecologia, 25, 5576.
  • Gibson, C.W.D. (1980) Niche use patterns among some Stenodemini (Heteroptera: Miridae) of limestone grasslands, and an investigation of the possibility of interspecific competition between Notostira elongata Geoffroy and Megaloceraea recticornis Geoffroy. Oecologia, 47, 352364.
  • Hägele, B.F. & Rowell-Rahier, M. (1999) Dietary mixing in three generalist herbivores: nutrient complementation or toxin dilution? Oecologia, 119, 521533.
  • Hector, A., Schmid, B., Beierkuhnlein, C., Caldeira, M.C., Diemer, M., Dimitrakopoulos, P.G. et al. (1999) Plant diversity and productivity experiments in European grasslands. Science, 286, 11231127.
  • Hoffman, G.D. & McEvoy, P.B. (1986) Mechanical limitations on feeding by meadow spittlebugs Philaenus spumarius (Homoptera: Cercopidae) on wild and cultivated host plants. Ecological Entomology, 11, 415426.
  • Houba, V., Van Vark, W., Walinga, I. & Van Der Lee, J.J. (1989) Plant Analysis Procedures. Department of Soil Science and Plant Analysis, Wageningen, The Netherlands.
  • Huston, M.A. (1997) Hidden treatments in ecological experiments: re-evaluating the ecosystem function of biodiversity. Oecologia, 110, 449460.
  • Kullenberg, B. (1944) Studien über die Biologie der Capsiden. PhD thesis, University of Uppsala, Sweden.
  • Mattson, W.J. (1980) Herbivory in relation to plant nitrogen content. Annual Review of Ecology and Systematics, 11, 119161.
  • McNeill, S. (1971) The energetics of a population of Leptopterna dolabrata (Heteroptera: Miridae). Journal of Animal Ecology, 40, 127140.
  • McNeill, S. (1973) The dynamics of a population of Leptopterna dolabrata (Heteroptera: Miridae) in relation to its food resources. Journal of Animal Ecology, 42, 495507.
  • McNeill, S. & Southwood, T.R.E. (1978) The role of nitrogen in the development of insect/plant relationships. Biochemical Aspects of Plant and Animal Coevolution (ed. by J. B.Harborne), pp. 7798. Academic Press, London.
  • Meister, E. & Lehmann, J. (1990) Leistungs- und Qualitätsmerkmale verschiedener Gräser bei steigender Stickstoffdüngung. Landwirtschaft Schweiz, 3, 125130.
  • Modder, W.W.D. & Tamu, G.F. (1996) The effect of food plants and metabolic reserves, development and fecundity in the African pest grasshopper, Zonocerus variegatus (Linnaeus) (Orthoptera: Pyrgomorphidae). African Entomology, 4, 189196.
  • Naeem, S., Thompson, L.J., Lawler, S.P., Lawton, J.H. & Woodfin, R.M. (1994) Declining biodiversity can alter the performance of ecosystems. Nature, 368, 734737.
  • Novozamsky, I., Houba, V.J.G., Van Eck, R. & Van Vark, W. (1983) A novel digestion technique for multi-element-analysis. Communications in Soil Science and Plant Analysis, 14, 239249.
  • Osborn, H. (1918) The meadow plant bug Miris dolabratus. Journal of Agricultural Research, 15, 175200.
  • Southwood, T.R.E. (1973) The insect/plant relationship – an evolutionary perspective. Symposium of the Royal Entomological Society of London, 6, 330.
  • Southwood, T.R.E. & Leston, D. (1959) Land and Water Bugs of the British Isles. Warne Ltd, London.
  • Strong, D.R., Lawton, J.H. & Southwood, T.R.E. (1984) Insects on Plants – Community Patterns and Mechanisms. Blackwell Scientific Publications, Oxford.
  • Tilman, D., Wedin, D. & Knops, J. (1996) Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature, 379, 718720.
  • Wardle, D.A. (1999) Is ‘sampling effect’ a problem for experiments investigating biodiversity–ecosystem function relationships? Oikos, 87, 403407.
  • White, T.C.R. (1993) The Inadequate Environment – Nitrogen and the Abundance of Animals. Springer Verlag, Berlin.

Accepted 29 July 2002