SEARCH

SEARCH BY CITATION

Keywords:

  • effect testing;
  • environmental risk assessment;
  • genetically modified plant;
  • population biological models

Summary

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

1.  The probability of a transgenic crop establishing a feral population outside cultivated areas and possibly outcompeting naturally occurring species needs to be assessed to make an ecological risk assessment of the transgenic crop.

2.  The interaction between herbivory and competition is thought to determine the ecological success of insect-resistant plants, and this interaction was investigated in a competition experiment with transgenic insect-resistant Bt-Brassica napus, Brassica rapa, Lolium perenne, and herbivory from the large white butterfly Pieris brassicae.

3.  As expected, herbivory had a negative effect on the biomass of B. rapa at high plant densities. The competitive ability of L. perenne, when growing with B. rapa, increased significantly with the level of herbivory on B. rapa.

4.  To predict the effect of herbivory in a natural ecosystem, plant competition between the two annual Brassica species was analysed in a population ecological model. It was concluded that it is probable that transgenic Bt-B. napus plants may invade a natural habitat if herbivory is sufficiently high and the habitat is suitable for B. napus.

5.Synthesis and applications. The results indicate that it is important to study the interaction between herbivory and competition when assessing the ecological risk of insect-resistant genetically modified crops. Furthermore, combining ecological data from manipulated experiments with population ecological modelling is a fruitful approach when conducting environmental risk assessments.


Introduction

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

Subsequent to the introduction of genetically modified plants (GMP) in the early nineties, a debate of the risks associated with the new technology and the consequent necessary regulation ensued. At present, it is commonly agreed that a scientific risk assessment should form the basis for a rational decision for each GMP, i.e. case by case assessment. Generally, it is also accepted that the most efficient method is to apply a tiered approach, meaning that relatively simple assessments are used to evaluate whether effects are possible. If this simple approach suggests that a certain risk is possible, then a more demanding and realistic testing procedure is conducted at the next tier to determine whether the risk is probable (Poppy 2000; Andow & Hilbeck 2004).

The rationale of the paper is to study the ecological processes that are relevant for assessing the ecological risk that a transgenic crop establishes a feral population outside cultivated areas, which may, possibly, lead to the local extinction of naturally occurring species. The study is intended to balance the need for simplicity with the need for ecological relevance. We have chosen a case with the insect-resistant Bt-Brassica napus as a model system, as it is a biologically complex situation, where the amount of herbivory may determine the competitive success of the transgenic plant in natural habitats (Vacher et al. 2004). Furthermore, this situation includes the possibility of between-species gene flow (Mikkelsen, Andersen & Jørgensen 1996; Hansen, Siegismund & Jørgensen 2001; Warwick et al. 2008).

Both competition from neighbouring plants as well as herbivory is known to have major negative impacts on growth, reproduction and survival of plants, and they are both believed to have the potential for structuring plant communities (e.g. Crawley 1989; Gurevitch, Morrison & Hedges 2000). Plant resistance to herbivorous insects (e.g. by producing Bt-endotoxins) may increase the competitive ability of the resistant plant compared with non-resistant isogenic counterparts (Stewart et al. 1997; Snow et al. 2003). Field observations suggest that the effects of competition and herbivory might not be independent, i.e. the effects of a given level of herbivory are expected to depend on the competitive effects of the neighbouring plants (Fowler & Rausher 1985; Gurevitch et al. 2000; Hambäck & Beckerman 2003), and it has also been asserted that invertebrate herbivores have a larger effect on their host plants when the plants are simultaneously subjected to other forms of stress such as competition or disturbance (Cottam, Whittaker & Malloch 1986; Crawley 1989; Rodríguez & Brown 1998).

The possible interaction between competition and herbivory may arise because of size-asymmetric competition for limiting resources among plants (e.g. Weis & Hochberg 2000). At low plant densities, individual plants may recover quickly from herbivory, but when growing together in dense mixtures, the plants subjected to herbivory would lose their position in the height hierarchy and be overtopped. Lee & Bazzaz (1980) investigated the effect of defoliation and competition for light on seed and biomass production in velvetleaf, Abutilon theophrasti. In their low-density treatment, where each plant was fully exposed to light, even 75% defoliation had no significant effect on reproduction. At high density, however, 75% defoliation reduced fecundity to almost 50%. Additionally, resistance against herbivory has been suggested to be associated with a general cost (Bergelson & Purrington 1996). Theoretically, such a cost of resistance should be more severe in stressful environments and, hence, increase when competition is intense (Bergelson & Purrington 1996). The aim of the study is to investigate the effect of the larvae of the large white butterfly Pieris brassicae on the competitive ability and ecological success of three plant species:

  • 1
     Bt-B. napus (‘oilseed rape’, strain W45), which has been genetically modified to become resistant against the larvae by inserting a gene that produces synthetic Bacillus thuringiensis endotoxin (Bt Cry1Ac; Stewart et al. 1996). The endotoxin specifically acts against butterflies and moths (Parker et al. 2000), and when a larvae of the large white butterfly feeds on leaves with the Cry1Ac endotoxin, it dies (C. Kjær, unpublished data).
  • 2
    Brassica rapa, an annual weed and close relative of B. napus, which is known to be affected by the large white butterfly.
  • 3
    Lolium perenne, a perennial grass that belongs to a different functional group than the two Brassica species, i.e. L. perenne has a different phenology and life history and differs with respect to the general ecological requirements compared with the two Brassica species (Grime, Hodgson & Hunt 2007). Both L. perenne and B. rapa are found in the same ruderal habitat types (Grime et al. 2007). Lolium perenne is not attacked by the larvae.

Our hypotheses were that increased herbivory will (i) reduce the biomass of B. rapa at all plant densities, but because of competition mainly when B. rapa was planted at a high density; (ii) reduce the competitive effect of B. rapa on Bt-B. napus and L. perenne; and (iii) augment the competitive effect of Bt-B. napus and L. perenne on B. rapa, i.e. the competitive response on B. rapa is increased.

The reason for including L. perenne in the experiment was to evaluate the effect of the herbivore on the ecological success of the transgenic insect-resistant crop in a real ecological setting. We suggest that it is insufficient only to compare the insect-resistant plant with non-resistant close relatives. The insect-resistant crop will compete with a whole suite of species, some of which will belong to different functional types that may be limited by other resources than the transgenic crop. Furthermore, it may also be important to compare the insect-resistant plant species with a species with a different life-history strategy (Stewart et al. 1997).

In addition to the conventional ecological investigation of the interaction between competition and herbivory among the four studied species, the results of the study will be used to quantify the probability that Bt-B. napus will invade and, possibly, outcompete other species in a natural habitat. This ecological scenario represents a case that is relevant in the ecological risk assessment of transgenic Bt-B. napus.

Materials and methods

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

Plants

A response surface competition experiment, where both densities and proportions were manipulated (Inouye 2001; Damgaard 2008), was conducted with three plant species at three insect herbivore densities. The three plant species used in the experiment were: (i) a transgenic cultivar of B. napus L. var Westar (W45), which had been genetically modified to express the Bt-toxin Cry1Ac (Stewart et al. 1996). When the larvae of the large white butterfly encounter a transgenic Bt-B. napus, they will stop eating within a few hours and die (unpublished data); (ii) B. rapa L., a susceptible host plant for the large white butterfly; and (iii) Lolium perenne L. a perennial grass species that is not a host for the large white butterfly. In all treatments involving the susceptible B. rapa plants, a parallel series of treatments were established where one or two larvae, respectively, of the large white butterfly were put on each plant.

Insects

A laboratory stock of the large white butterfly Pieris brassicae L. was reared on B. napus var. Iris. The adult butterflies were kept in large cages (W × L × H = 70 × 100 × 100 cm) in a greenhouse. The butterflies were supplied with 10% sucrose water in artificial flowers as well as B. napus plants for oviposition. Egg masses for experiments were collected and placed in humid Petri dishes. The newly hatched larvae were used in the experiments.

Competition experiment

The three plant species were grown either in monoculture or in a one to one mixture of two of the species (B. rapa × Bt-B. napus, B. rapa × L. perenne and Bt-B. napus × L. perenne) at three densities (60, 90 and 140 plants m−2). The densities were chosen so as to span the range from crowded conditions to ample plant space. To balance the number of individual plants of the different density treatments, a fixed number of plants from each species was placed in boxes with variable area, i.e. the boxes with plants at low density were larger than the boxes with plants at high density. Each monoculture treatment included 50 plants, and each mixture treatment included 100 plants (50 of each species).

The above competition experiment was repeated at three levels of herbivory for all treatments which included susceptible B. rapa plants, and either 0, 1 or 2 larvae were placed on all the susceptible B. rapa plants.

The seeds were sown in planting-trays filled with humid potting-soil and covered with black plastic. The trays were stored at 5 °C for 4 days to ensure synchronized germination. Hereafter, the planting-trays were placed in a greenhouse and the cover was removed. Seedlings (8–12 days old) were transplanted and randomly assigned to a specific position in the experimental boxes to minimize the spatial covariance of the genotypes (Damgaard 2004b). The boxes were placed in a greenhouse with 12-h photoperiods at an average temperature of 20 °C (minimum: 14 °C and maximum: 37 °C) and a relative humidity of 64% (minimum: 48% and maximum: 83%). The boxes were placed close together to minimize edge effects and filled with a former agricultural soil (sandy loam), which was characterized by low levels of nutrients. The plants were watered from the bottom with tap water without added fertilizers or pesticides. Larvae were added 2 weeks after transplantation and placed on the lower rosette leaves. The larvae were not artificially restricted to feed on a single plant, but observations prior to and during the experiment showed that larvae mostly stayed on the plant that they were placed on. In those cases where a larva moved from a B. rapa-plant to a transgenic Bt-B. napus, it took a few bites, stopped feeding and eventually died. The experiment was terminated when the majority of the plants had reached the late reproductive stage (50 days). The shoot biomass of all plants was harvested, dried at 80 °C for 24 h, and weighed individually.

The observed variation in biomass may arise from three different sources: among plants within a plot, among plots and among treatments. However, based on our experience with observing individual competing plants, which often display a skewed size-distribution because of size-asymmetric growth and a sizeable border effect, and as there were logistic constraints on the number of experimental plants, the number of plants within a plot was augmented in the experimental design at the price of measuring a possible among-plot effect. That is, the among-plot variation was a priori assumed to be negligible compared with the two other sources of variation and the treatments were not replicated. This assumption was validated (i) in a pilot experiment where the among-plot variation in the seed production of B. napus was <1% of the variation observed within-plots and (ii) by visual assessments of the plots during the course of the experiment. Furthermore, as the treatment effects were analysed in a regression-based response surface competition model, where the primary goal is to determine the shape of the response surface, a possible among-plot effect among neighbouring treatment points will be somewhat ‘averaged out’ (Inouye 2005).

Competition model

The effect of herbivory on biomass of the three species was modelled by a generalization of a discrete hyperbolic competition model (Damgaard 2003, 2004a; Damgaard, Mathiassen & Kudsk 2008). We assume that, within the limited domain of the conducted competition experiment, herbivory affects the competitive interactions and the biomass of the susceptible B. rapa plants linearly, i.e.:

  • image( eqn 1)

where the indices indicate the species (1: Bt-B. napus, 2: B. rapa, 3: L. perenne), Yi is the biomass of species i, Xi is the density of species i, di and fi, which both are positive, are shape parameters of the response function of genotype i and assumed to be independent of the herbivory level. (1/ai)fi is a measure of the biomass of a plant of species i at low density in a monoculture in the absence of herbivory and α measures the effect of herbivory on B. rapa. If di = fi = 1, then 1/bi is a measure of the biomass of species i per area at high density in a monoculture in the absence of herbivory, and β measures the effect of herbivory on B. rapa. The competition coefficient of species j on species i without herbivory is measured by cij, and γij measures the effect of herbivory on the competition coefficients. The assumptions underlying the empirical competition model and the meaning of the parameters are further explained in Appendix S1 (Supporting Information).

The model was fitted to the treatment effects after both the data and the competition model were Box-Cox-transformed (λ1 = −1·5, λ2 = −1·5) (Seber & Wild 1989). After transformation, the residuals were approximately normally distributed with a homogeneous variance. To avoid negative parameter values, the simple functions involving ai, α, bi, β, di and fi were transformed with the exponential function, e.g. a[RIGHTWARDS ARROW] exp(a1), a2 + αh [RIGHTWARDS ARROW] exp(a2 + αh).

The competition model (1) is quite flexible, and, in many cases, the model will be over-parameterized. Such a possible over-parameterization, generally, reduces the testing power of the model. Consequently, it was tested whether the competition model could be simplified by setting di and fi equal to one using a likelihood ratio test.

The joint Bayesian posterior distribution of the parameters in the competition model was sampled using the Metropolis-Hastings algorithm with a multinomial candidate distribution (100 000 iterations with a burn-in period of 1000) assuming uniform prior distributions of the location parameters. The sampling procedure was checked by visual inspections of the sampling chains as well as by computing the autocorrelation and acceptance ratio (Carlin & Louis 1996). Statistical inferences on the parameters were made using the posterior distribution of the parameters, i.e. the 95% credibility level of the parameter.

Results

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

The dry weight data of the three species are shown in Fig. 1 with the mean and standard error for each treatment. As expected, there is a clear negative effect of plant density, and, as the effect of density on biomass is known to be nonlinear (e.g. Damgaard 2004a), the biomass data were analysed using the competition model (1). Using likelihood ratio tests, it was observed that the shape parameters fi and di were not significantly different from one (P = 0·45 and 0·39 respectively), and the competition model (1) was, consequently, simplified by setting these two parameters to one.

image

Figure 1.  Biomass of the three investigated plant species under different biotic conditions: competing species, plant density and level of herbivory. The left column presents biomass of Brassica rapa competing with either Lolium perenne or Bt-Brassica napus, or growing in monoculture with different levels of herbivore load (0, 1 or 2 Pieris brassicae larvae). The upper right graph presents biomass of the transgenic B. napus in monoculture or in competition with either B. rapa at different levels of herbivory (Br-herb 0, Br-herb 1 or Br-herb2) or Lolium. The lower right graph presents the biomass of Lolium perenne in monoculture or competing with either B. rapa at different levels of herbivory or Bt-B. napus. Density is the joint density of all species present in the experimental setup.

Download figure to PowerPoint

The competitive ability of B. rapa without herbivores and of L. perenne was high compared with Bt-B. napus, i.e. competition coefficients were significantly larger than one (Table 1), whereas B. rapa without herbivores was relatively unaffected by competition from either Bt-B. napus or L. perenne, i.e. the competition coefficients were significantly lower than one (Table 1). The same was true for L. perenne, which was relatively unaffected by either B. rapa without herbivores or Bt-B. napus.

Table 1.   95% Credibility intervals and the median (50%) of the simulated marginal Bayesian posterior distribution of the parameters in competition model (1)
Parameter2·50%50%97·50%
  1. The indices indicate the species (1, Bt-Brassica napus; 2, B. rapa; 3, Lolium perenne).

 a1−12·164−9·120−7·140
 a2−5·260−4·552−3·588
 a3−4·127−3·187−2·744
 b1−5·852−5·753−5·642
 b2−6·130−6·001−5·894
 b3−6·408−6·213−6·030
Competition coefficient
 c121·4621·8242·323
 c131·1711·6042·090
 c210·2030·4030·606
 c230·4000·6610·907
 c310·2300·5080·848
 c320·0530·2820·552
Effect of herbivory
 α−0·855−0·2010·493
 β0·2240·3090·400
 γ12−0·2920·0250·300
 γ21−0·225−0·0790·078
 γ230·4440·8171·252
 γ32−0·186−0·0640·055

Herbivory affected both biomass at high plant density and the competitive interactions; two of the six parameters that measure effect of herbivory were significantly different from zero (Table 1); (i) β that measures the effect of herbivory on the biomass of B. rapa at high density, was significantly larger than zero, indicating that the biomass, as expected, decreased with increasing level of herbivory. However, α did not differ significantly from zero, indicating a limited effect of the herbivore when B. rapa grew at low densities, and (ii) γ23 that measures the effect of herbivory on the competitive ability of L. perenne when growing together with B. rapa was significantly larger than zero, indicating that B. rapa was more affected by L. perenne when the level of herbivory increased. The other three parameters that measure the effect of herbivory on competitive ability (γ12,γ21 and γ32) were not significantly different from zero.

Using the empirical competition model (1), it was demonstrated that the larvae of the large white butterfly had negative effects on both biomass and the competitive ability of B. rapa, but the effect of the herbivore on the dynamics in plant communities with B. rapa was relatively unclear. To assess the long-term ecological significance of the herbivore, we predicted the long-term effect of herbivory in a population ecological model of plant competition among the two annual Brassica species (Damgaard 2003) using empirical regression models of the relationship between biomass and the number of produced seeds (unpublished data) and assumptions on the density-independent probability of germination and early establishment (Damgaard 2004a). If it is assumed that the probability of germination and early establishment for both Bt-B. napus and B. rapa is equal to 1%, then the probability that Bt-B. napus will invade a natural habitat is predicted to be significant when the average number of larvae per plant is higher than two (Fig. 2). Furthermore, the probability that Bt-B. napus will outcompete B. rapa is predicted to be significant when the average number of larvae per plant is higher than three (Fig. 2). However, the number of herbivores per plant in the competition experiment was either 0, 1 or 2, and any predictions outside the domain of the experiment should, of course, be met with caution. These results were only moderately sensitive to the assumed probability of germination and early establishment in such a way that, if the probabilities were lowered for both species, then a slight increase in herbivory was required for Bt-B. napus to outcompete B. rapa (results not shown).

image

Figure 2.  The predicted probabilities of the different ecological scenarios when a transgenic herbivore-resistant Bt-Brassica napus and B. rapa compete as functions of the average number of larvae per plant using the methodology described in Damgaard (2003) and setting probability of germination and early establishment = 0·01. ‘Brassica rapa’ denotes the scenario where B. rapa outcompetes Bt-B. napus, ‘B. napus’ denotes the scenario where Bt-B. napus outcompetes B. rapa, and ‘Coex’ denotes the scenario where the two species will co-exist at equilibrium. The number of herbivores per plant in the experiment was 0, 1 and 2, and the stippled line marks the domain of the experiment. Note that the level of herbivory, where B. rapa is predicted to be outcompeted by the transgenic rapeseed, is outside the domain of the experiment, and these predictions should, therefore, be met with some caution.

Download figure to PowerPoint

Discussion

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

Generally, Bt-B. napus was not found to be competitive when growing together with naturally occurring species. This finding is in agreement with several studies, where winter survival of B. napus in naturally vegetated plot has been found to be very sparse (Crawley et al. 1993; Stewart et al. 1997; Crawley et al. 2001).

As expected, herbivory had a negative effect on the biomass of B. rapa at high plant densities, whereas there was no effect at low plant densities. This result corroborates the hypothesis that herbivory has the largest effect when competition is strong (e.g. Weis & Hochberg 2000). Furthermore, the competitive ability of L. perenne towards B. rapa increased significantly with the level of herbivory. However, we did not observe a significant change in the competitive ability of B. rapa towards L. perenne, and the competitive interactions between Bt-B. napus and B. rapa were not significantly affected by herbivory at the intermediate level of herbivory used in this experiment. However, these findings should be contrasted with the conclusions of Stewart et al. (1997) based on manipulated field plots, where medium to high levels of defoliation decreased survivorship of non-transgenic B. napus plants relative to Bt-B. napus and increased reproduction in favour of Bt-B. napus.

Initially, we speculated that the interaction between competition and herbivory between two species would be stronger if the two species belong to the same functional type because of competition for the same limiting resource (Silvertown 2004). However, the only significant effect of herbivory on the competitive interactions was observed with L. perenne, which belongs to a different functional group than B. rapa, and not with Bt-B. napus, which belongs to the same functional group as B. rapa. Thus, this hypothesis was not supported by the results, possibly because light, which is a critical resource for both L. perenne and B. rapa, may have been the most important limiting resource in the manipulated experiment.

The larvae in this experiment were distributed evenly among the B. rapa plants. However, in a natural setting the female butterfly will deposit up to 150 eggs on a single plant and many of the surrounding plants will be free of eggs. After the larvae hatch and have defoliated the plant, they will migrate to the surrounding plants and start to feed on them. Consequently, it is difficult to compare the degree of herbivory in our manipulated experiment, where all B. rapa plants received a few larvae at the same time, with the degree of herbivory experienced in natural settings. Kristensen (1994) followed the survival of egg clutches (7–105 eggs per clutch with an average of 50·2) from hatch to pre-pupae in an agricultural setting. He found that 50% died in each larval stage meaning that only 3% survived to adult stage because of predation, parasitism and other causes. In another field experiment, there was <1 P. brassica-larva on average per plant in 30 different patches (Grez & Gonz lez 1995). Therefore, the degree of herbivory in a natural setting corresponds to a degree of herbivory between one and five larvae in our manipulated experiment. However, more work needs to be performed on the linkage between the two settings, and we have initiated experiments and modelling studies to answer this question.

The present experiment was conducted with a sufficiently low number of larvae so that they did not have to leave the plant to complete larval development. Under these circumstances, the experiment is representative of other butterfly species feeding on B. napus, such as Plutella xylostella and P. rapa, that adjust the number of eggs to the plant size.

To assess whether genetically modified insect-resistant Bt-B. napus plants might invade a natural habitat, we modelled the effect of herbivory on plant competition among the two annual Brassica species, and we concluded that this was probable if the number of larvae of the large white butterfly per plant was sufficiently high (Fig. 2) and the habitat was suitable for B. napus, such as disturbed or ruderal habitats (Crawley & Brown 1995; Pessel et al. 2001). However, as mentioned above, at present it is unclear whether the required amount of herbivory is realistic in a natural ecological setting, and more information on the expected probability of germination and early establishment of Bt-B. napus and B. rapa in similar conditions is needed. In a pilot study, it was found that when there are c. 4–5 larvae on each individual plant then the plant is completely eaten in a relatively short time and the larvae leave the plant. Thus, the predicted herbivore density of 3–5 larvae per plant, which is expected to have effects on plant community structures, is outside the possible domain of the manipulated experiment. Consequently, the manipulated experiment in a greenhouse may be viewed as a lower tiered approach in a comprehensive risk assessment of Bt-B. napus and, to make a reliable ecological risk assessment of the probability that the insect-resistant Bt-B. napus establishes a feral population outside cultivated areas, it is necessary to move to the next tier in the risk assessment procedure (Poppy 2000; Andow & Hilbeck 2004), and investigate the relevant ecological processes in a semi-field or field experiment. Furthermore, the complicating effect of gene flow between transgenic B. napus and B. rapa may be considered as a separate and independent ecological hazard (Mikkelsen et al. 1996; Hansen et al. 2001; Warwick et al. 2008).

In a scientifically based risk assessment, the magnitude of a risk is measured as a function of the probability that a specific event will occur and the negative impact of the event (e.g. Vose 2000). This definition of risk has been adopted by the EU Commission in the procedure for ecological risk assessment of GMP (EU, 2001, 2002). However, the actual ecological risk assessment of GMP in the EU countries has, until now, mainly been qualitative and based on expert opinions, although quantitative ecological risk assessment procedures have been proposed for different types of ecological risks (e.g. Damgaard & Løkke 2001; Damgaard 2002). This study considers how to proceed towards making a more comprehensive and scientific assessment of the complex ecological risk that an insect-resistant GMP may invade a natural habitat.

Acknowledgements

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

This study was performed within the framework of the project ‘Biotechnology: elements in environmental risk assessment of genetically modified plants’. Seeds of transgenic canola as well as the parent canola were kindly supplied by Neil Stewart. The authors thank Inger Møller, Lise Lauridsen and Trine Guldager Sørensen for practical assistance, and Charlotte Kler, Gösta Kjellsson and Vibeke Simonsen for commenting on a previous version of the manuscript.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Andow, D.A. & Hilbeck, A. (2004) Science-based risk assessment for nontarget effects of transgenic crops. BioScience, 54, 637649.
  • Bergelson, J. & Purrington, C.B. (1996) Surveying patterns in the cost of resistance in plants. American Naturalist, 148, 536558.
  • Carlin, B.P. & Louis, T.A. (1996) Bayes and Empirical Bayes Methods for Data Analysis. Chapman & Hall, London.
  • Cottam, D.A., Whittaker, J.B. & Malloch, A.C.J. (1986) The effects of Chrysomelid bettle grazing and plant competition on the growth of Rumex obtusifolius. Oecologia, 70, 452456.
  • Crawley, M.J. (1989) Insect herbivores and plant population dynamics. Annual Review of Entomology, 34, 531564.
  • Crawley, M.J. & Brown, S.L. (1995) Seed limitation and the dynamics of feral oilseed rape on the M25 moterway. Proceedings of the Royal Society of London, Series B, 259, 4954.
  • Crawley, M., Hails, R.S., Rees, M., Kohn, D. & Buxton, J. (1993) Ecology of transgenic oilseed rape in natural habitats. Nature, 363, 620622.
  • Crawley, M.J., Brown, S.L., Hails, R.S., Kohn, D.D. & Rees, M. (2001) Transgenic crops in natural habitats. Nature, 409, 682683.
  • Damgaard, C. (2002) Quantifying the invasion probability of genetically modified plants. BioSafety, 7, Paper 1 (BY02001) (Online Journal). Available at: http://www.bioline.org.br/by.
  • Damgaard, C. (2003) Modelling plant competition along an environmental gradient. Ecological Modelling, 170, 4553.
  • Damgaard, C. (2004a) Evolutionary Ecology of Plant–Plant Interactions – An Empirical Modelling Approach. Aarhus University Press, Aarhus.
  • Damgaard, C. (2004b) Inference from plant competition experiments: the effect of spatial covariance. Oikos, 107, 225230.
  • Damgaard, C. (2008) On the need for manipulating density in competition experiments. Functional Ecology, 22, 931933.
  • Damgaard, C. & Løkke, H. (2001) A critique of the “concept of familiarity” as used in the ecological risk assessment of genetically modified plants. BioSafety, 6, Paper 1 (BY01001) (Online Journal). Available at: http://www.bioline.org.br/by.
  • Damgaard, C., Mathiassen, S.K. & Kudsk, P. (2008) Modelling effects of herbicide drift on the competitive interactions between weeds. Environmental Toxicology and Chemistry, 27, 13021308.
  • EU (2001) Directive 2001/18/EC of the European Parliament and of the Council of 12 March 2001 on the deliberate release into the environment of genetically modified organisms and repealing Council Directive 90/220/EEC. Official Journal of the European Communities, 106, 139.
  • EU (2002) 2002/623/EC: Commission Decision of 24 July 2002 establishing guidance notes supplementing Annex II to Directive 2001/18/EC of the European Parliament and of the Council on the deliberate release into the environment of genetically modified organisms and repealing Council Directive 90/220/EEC (Text with EEA relevance) [notified under document number C(2002) 2715]. Official Journal of the European Communities, 200, 2233.
  • Fowler, N.L. & Rausher, M.D. (1985) The joint effect of competitors and herbivores on growth and reproduction in Aristlochia retculata. Ecology, 66, 15801587.
  • Grez, A.A. & Gonz lez, R.H. (1995) Resource concentration hypothesis: effect of host plant patch size on density of herbivorous insects. Oecologia, 103, 471474.
  • Grime, J.P., Hodgson, J.G. & Hunt, R. (2007) Comparative Plant Ecology, 2nd edn. Castlepoint Press, Colvend.
  • Gurevitch, J., Morrison, J.A. & Hedges, L.V. (2000) The interaction between competition and predation: a meta-analysis of field experiments. American Naturalist, 155, 435453.
  • Hambäck, P.A. & Beckerman, A.P. (2003) Herbivory and plant resource competition: a review of two interacting interactions. Oikos, 101, 2637.
  • Hansen, L.B., Siegismund, H.R. & Jørgensen, R.B. (2001) Introgression between oilseed rape (Brassica napus L.) and its weedy relative B. rapa L. in a natural population. Genetic Resouces and Crop Evolution, 48, 621627.
  • Inouye, B.D. (2001) Response surface experimental designs for investigating interspecific competition. Ecology, 82, 26962706.
  • Inouye, B.D. (2005) The importance of the variance around the mean effect size of ecological processes: comment. Ecology, 86, 262265.
  • Kristensen, C.O. (1994) Investigations on the natural mortality of eggs and larvae of the large white Pieris brassicae (L.) (Lep., Pieridae). Journal of Applied Entomology, 17, 9298.
  • Lee, T.D. & Bazzaz, F.A. (1980) Effects of defoliation and competition on growth and reproduction in the annual plant Abutilon theophrasti. Journal of Ecology, 75, 871886.
  • Mikkelsen, T.R., Andersen, B. & Jørgensen, R.B. (1996) The risk of crop transgene spread. Nature, 380, 31.
  • Parker, C.D.J., Mascarenhas, V.J., Luttrell, R.G. & Knighten, K. (2000) Survival rates of tobacco budworm (Lepidoptera: Noctuidae) larvae exposed to transgenic cottons expressing insecticidal protein of Bacillus thuringiensis Breliner. Journal of Entomological Science, 35, 105117.
  • Pessel, F.D., Lecompte, J., Emeriau, V., Krouti, M., A., M. & Gouyon, P.H. (2001) Persistence of oilseed rape (Brassica napus L.) outside of cultivated fields. Theroretical and Applied Genetics, 102, 841846.
  • Poppy, G. (2000) GM crops: environmental risks and non-target effects. Trends in Plant Science, 5, 46.
  • Rodríguez, M.A. & Brown, V.K. (1998) Plant competition and slug herbivory: effects on the yield and biomass allocation patterns of Poa annua L. Acta Oecologica, 19, 3746.
  • Seber, G.A.F. & Wild, C.J. (1989) Nonlinear Regression. John Wiley & Sons, New York.
  • Silvertown, J. (2004) Plant coexistence and the niche. Trends in Ecology and Evolution, 19, 605611.
  • Snow, A.A., Pilson, D., Rieseberg, L.H., Paulsen, M.J., Pleskac, N., Reagon, M.R., Wolf, D.E. & Selbo, S.M. (2003) A Bt transgene reduces herbivory and enhances fecundity in wild sunflowers. Ecological Applications, 13, 279286.
  • Stewart, C.N.J., Adang, M.J., All, J.N., Raymer, P.L., Ramachandran, S. & Parrott, W.A. (1996) Insect control and dosage effect in transgenic canola containing a synthetic Bacillus thuringiensis cryIAc gene. Plant Physiology, 112, 115120.
  • Stewart, C.N.J., All, J.N., Raymer, P.L. & Ramachandran, S. (1997) Increased fitness of transgenic insecticidal rapeseed under insect selection pressure. Molecular Ecology, 6, 773779.
  • Vacher, C., Weis, A.E., Hermann, D., Kossler, T., Young, C. & Hochberg, M.E. (2004) Impact of ecological factors on the initial invasion of Bt transgene into wild populations of birdseed rape (Brassica rapa). Theoretical and Applied Genetics, 109, 806814.
  • Vose, D. (2000) Risk Analysis – A Quantitative Guide, 2 edn. John Wiley & sons, Chichester.
  • Warwick, S.I., Légère, A., Simard, M.-J. & James, T. (2008) Do escaped transgenes persist in nature? The case of an herbicide resistance transgene in a weedy Brassica rapa population. Molecular Ecology, 17, 13871395.
  • Weis, A.E. & Hochberg, M.E. (2000) The diverse effects of intraspecific competition on the selective advantage to resistance: a model and its predictions. American Naturalist, 156, 276292.

Supporting Information

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

Appendix S1. Assumptions of the competition model.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
JPE_1689_sm_AppendixS1.doc42KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.