• biological control;
  • enemy release hypothesis;
  • insect–plant interactions;
  • invaded–native range comparison;
  • matrix projection models;
  • special herbivores;
  • vital rate elasticity


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    The invasive thistle Carduus nutans causes major economic losses in the Americas, Australia and New Zealand. For the first time, we have modelled its population dynamics in its native range, Eurasia, where it rarely reaches problematic densities, in order to identify ways to improve management strategies for this weed in the invaded range.
  • 2
    In particular, we investigated whether specialist enemies in the thistle's native range suppress thistle populations, as predicted by the enemy-release hypothesis, and, if so, how this effect relates to other factors that may limit population growth.
  • 3
    We constructed population transition matrix models with data from three French populations. A vital rate elasticity analysis revealed that reproduction determines between 33% and 61% of the projected growth rate of the populations, and thus is a key driver of the population dynamics of this monocarpic short-lived perennial.
  • 4
    Decreases in population size were predicted by the models for all three populations (λ < 1). Using limiting factor omission analyses, we showed that the suite of native insect herbivores causing seed losses had a larger impact than the joint effects of rosette damage by sheep and summer drought acting on seedling establishment. Removal of insect herbivores increased the native population growth rate by 166% on average; removal of sheep damage and summer drought from the model increased population growth by 51%. Specialist herbivores and drought interacted synergistically to affect reproduction.
  • 5
    Synthesis and applications. We show that vital rate elasticity analysis provides more management information than elasticity analysis on the level of matrix elements, particularly when management options affect only certain vital rates. For example, it can be used to make predictions about the effectiveness of predispersal floral herbivores in control management. The model can also help identify ways in which biocontrol can be augmented using integrated weed management that reduces seedling establishment probabilities, for example by preventing overgrazing. This method illustrates how native range studies of invasive species can be used to generate insights into managing populations in the invaded range.


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

The enemy-release hypothesis suggests that the success of invasive plant species in the invaded range may be explained by the absence of specialist herbivores (Keane & Crawley 2002; Colautti et al. 2004; DeWalt, Denslow & Ickes 2004; Carpenter & Cappuccino 2005). This is the key motivation behind introducing biological control agents. The success of biological invasions by plants, however, may not only depend on the (absence of) specialist natural enemies in the invaded ecosystem, but also on available resource niches, competition with neighbouring plants and the grazing intensity of generalist herbivores (Shea & Chesson 2002; van Ruijven, De Deyn & Berendse 2003; Huston 2004).

The Eurasian thistle Carduus nutans L. (Asteraceae), an invasive alien weed in the Americas, Australia and New Zealand, has been the target of biocontrol by several flower head- and rosette-feeding insects. However, biological control has had variable success in the invaded range (Woodburn 1993, 1997), as the effectiveness of the insects seems to depend strongly on local ecological conditions (Harris 1984; Kelly & McCallum 1992; Shea & Kelly 1998; Shea et al. 2005). The recent debate on the impact of floral herbivores (van Klinken et al. 2004) also implicitly calls for an analysis of how important predispersal seed survival is for the population growth rate of weeds.

According to the enemy-release hypothesis (Keane & Crawley 2002), specialist herbivores should be more abundant in native plant populations than in uncontrolled populations in the invaded range. Therefore, key insights into the role of natural enemies on the population dynamics of the plant, and their potential for biological control, could be gained from examination of the population dynamics in the native range. Most studies of the population dynamics of invasive plants in general (Lonsdale, Farrell & Wilson 1995; Buckley et al. 2004, 2005), and of C. nutans in particular (Lee & Hamrick 1983; Popay & Medd 1995; Shea et al. 2005; Shea, Sheppard & Woodburn 2006), have, however, focused on the invaded range. Native range studies, in this context, have been pioneered by Paynter et al. (1998) and Grigulis et al. (2001), and recent reviews have called for more such studies (Hinz & Schwarzlaender 2004; Hierro, Maron & Callaway 2005).

This study investigated the relative contributions of different life-cycle components and biotic and abiotic factors to the population dynamics of C. nutans in its native range, where it is a relatively scattered pasture species. We synthesized and used published and unpublished demographic data from three populations in southern France (Sheppard et al. 1989; Sheppard, Cullen & Aeschlimann 1994). To our knowledge, this is the only existing demographic data set from this species’ native range. We constructed projection matrix models and applied recently advocated vital rate elasticity analyses (Franco & Silvertown 2004; Morris & Doak 2004) to evaluate the relative importance of stage-specific impacts on the underlying vital rates and hence on the population dynamics of the native populations. Impacts assessed included flower head herbivory by insects, damage from sheep grazing and changes in seedling recruitment. This exploration of the factors limiting an invasive species in its native range augments our understanding of biocontrol impact in the invaded range (Shea et al. 2005) and suggests ways to improve control using integrated weed management.


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

study system

The nodding or musk thistle C. nutans is of Eurasian origin but in the last centuries has invaded the Americas, southern Africa, Australia and New Zealand, where it is a noxious weed in pastures and roadsides (Kelly & Popay 1985; Popay & Medd 1995). It is a monocarpic perennial plant and does not reproduce clonally. Bolting plants grow stems up to 2 m tall, on which capitula are formed through indeterminate growth. The wind-dispersed seeds can persist for at least a decade when buried in the soil (Burnside et al. 1981) but only a few years when soils are disturbed (Roberts & Chancellor 1979). Seedlings require open microsites for establishment and are sensitive to summer droughts (Medd & Lovett 1978; Wardle, Nicholson & Rahman 1992).

In order to model the population dynamics in the native range of this species, we used demographic data from Sheppard et al. (1989), Sheppard, Cullen & Aeschlimann (1994) and A. W. Sheppard et al. (unpublished data) from the Massif Central, southern France. This region was selected because its climatological conditions resemble those of the region of Australia where C. nutans is a major economic pest. Two of the three populations studied (populations 1 and 3) were in relatively infertile limestone depressions, while population 2 was in a more fertile and more actively managed pasture (Sheppard, Cullen & Aeschlimann 1994). The populations were well established at the beginning of the study (several thousands plants at each site; Table 1). All sites were used for grazing by sheep, and specialist herbivores were present at each site.

Table 1.  Description of the three Carduus nutans populations near La Cavalerie (Département Aveyron, France; Sheppard et al. 1995). The predispersal seed loss data are from Sheppard, Cullen & Aeschlimann (1994)
  Carduus nutans population
Soil fertilityLowHighLow
Population area1200 m2960 m2900 m2
Quadrat size1 × 0·5 m21 × 1 m21 × 1 m2
Number of quadrats203520
Density of plants (plants m−2)3·13·321·0
First census2 May 198619 June 198622 October 1987
Last census4 May 19896 October 19874 May 1989
Number of censuses392614
Predispersal seed loss to specialist herbivores (%)928491

matrix construction

To study the dynamics of the French populations and the relative effects of specialist herbivores, generalist herbivory and microsite limitation, we re-analysed the demographic data and constructed population projection matrices. Transition matrices, which contain the probabilities of survival, growth and contributions via reproduction from every discrete class of individual plants to each class in the next time step, are regularly used to analyse population dynamics (Caswell 2001). We used the year-to-year transition matrix structure of Shea & Kelly (1998) for C. nutans, recognizing four stage classes: SB, seeds in the seed bank; S, small plants; M, medium-sized plants; L, large plants. Size was based on the maximal rosette area in a year, which was estimated by measurements of the widest rosette diameter, d1, and its perpendicular rosette diameter, d2. Rosette area, a, was calculated with the following formula:

  • image(eqn 1)

The size class boundaries between the S and the M classes, and between M and L, were based on the bolting probabilities of the plants in these classes. Plants with 20% or less probability of bolting the next year were placed in the S category; plants with 80% or more probability of bolting were assigned to the L category; and the remainder were assigned to the M class (Shea & Kelly 1998; Shea et al. 2005). To determine the rosette area values of the class boundaries, we aggregated all data from the three populations and fitted the bolting probability, p(b), with a binary logistic regression model with all plants that survive until the next year (n = 630, P < 0·001):

  • image(eqn 2)

The boundary rosette areas that match p(b)-values of 0·2 and 0·8 were 26·3 cm2 and 60·7 cm2, respectively.

Survival, growth, bolting, flowering and seed production were studied in each population in at least 20 randomly placed permanent 1-m2 quadrats (Table 1; Sheppard et al. 1989; Sheppard, Cullen & Aeschlimann 1994). In population 1, all plants in these quadrats were monitored every 6 weeks, and fortnightly during the flowering season, from 1986 to 1989 (i.e. 3 year-to-year transitions) (Sheppard et al. 1989). Population 2 was monitored from 1986 to 1987 (one transition) until the site owner mowed the population (Woodburn & Sheppard 1996). Population 3 was studied from 1987 to 1989 (two transitions; A.W. Sheppard et al. unpublished data). For each population, demographic data from different years were lumped to increase sample sizes. For each size class in each population, we sequentially calculated the following vital rates: σj, the probability that a rosette or seed survived till the next year; βj, the probability that a surviving rosette bolted; γij, the probabilities that a surviving rosette or seedling (γi) grew into the different size classes. Bolting plants flowered and almost invariably died afterwards, whether they succeeded in flowering and seed production or not. Seed production was estimated by measuring the diameter of all flower heads that opened on every bolted plant. We distinguished four size classes of flower heads (diameter < 13, 13–20, 21–30 and > 30 mm). Average seed production without insect herbivores, πj (respectively 56, 172, 439 and 701 viable seeds for the four head size classes), was determined by spraying 57 heads on additional plants at site 2 with insecticides (Sheppard, Cullen & Aeschlimann 1994). These potential seed numbers were then compared with the seed production of unsprayed heads of similar flower head sizes in all three sites to determine the site-specific proportion of potential seeds surviving insect floral herbivory, ϕ.

A seed-addition experiment was conducted at sites 1 and 3 in December 1987 to estimate seedling establishment. In a factorial block design, 8 of 16 pasture plots of 0·125 m2 were each sown with 625 viable seeds mixed in a fine sand mixture. The establishment probability per seed, ɛ, was calculated as the number of seedlings that recruited during 1988 in the seeded plots (corrected for background germinations by subtracting the seedling numbers of unseeded plots) divided by the number of added seeds, and was on average 0·09 (SD = 0·06) and 0·11 (SD = 0·04) for sites 1 and 3, respectively. We assumed that seedling establishment was comparable (0·10) in population 2.

Seed survival in the soil was estimated from the same experiment using two 3·2-cm diameter × 10-cm deep soil cores per plot. The survival of new seeds was determined by counting the number of viable seeds in cores from seeded plots in July 1988 and October 1988. The proportion of the seeds that survived over this period was on average 0·32. This value was used in the matrix models as the probability that new seeds are incorporated into the seed bank (i.e. not germinating or dying), ν. As no data were available on the year-to-year survival in the seed bank, σ1, by seeds older than 1 year, we used the proportion of seeds that were still viable 1 year later (July 1989) in the seed-addition experiment as an estimate. This value was on average 0·26. In summary, each vital rate was represented by a symbol (Fig. 1 lists all vital rates and symbols) and groups of vital rates comprised the elements of the 4 × 4 (SB, S, M and L) annual transition matrix:


Figure 1. Elasticity values of all vital rates in the dynamics of three populations of Carduus nutans. For each matrix the elasticity values were rescaled in such a way that their absolute values summed to 1, in order to allow for direct comparison with the elasticity values of matrix elements presented in Fig. 2. The vital rates are divided into three groups: reproduction, survival and growth (stasis or size changes of surviving plants). All vital rates are proportions, except for the potential seed production rates. SB, seed bank; S, small plants; M, medium-sized plants; L, large plants. See the Methods for an explanation of how the vital rates were estimated and equation 3 for how the vital rates combine to form the elements of a matrix model.

Download figure to PowerPoint

  • image(eqn 3)

Each of the 16 matrix elements aij represented the average contribution (through survival and growth or through reproduction) of an individual in that stage class in year t (column j) to the number of individuals in a stage class in the next (t + 1) year (row i).

matrix analysis

We calculated the projected population growth rates (λ, the dominant eigenvalue) for the three constructed 4 × 4 matrices. A λ-value below unity means that a population will decline in size if the vital rates stay constant through time. To estimate the uncertainty (95% confidence intervals) of each λ-value, we resampled all underlying data sets 3000 times using the bootstrap method (Efron 1982; Kalisz & McPeek 1992). In order to predict the impact of different biotic and abiotic factors, such as floral herbivory and drought, on λ, we quantified the importance of the life-history components that were affected by these factors. First, we evaluated how much each of the 16 matrix elements contributed to λ by calculating the elasticity (eij, the relative sensitivity) of λ to each element, i.e. the proportional change in λ in response to small, proportional perturbations of each element separately (de Kroon et al. 1986):

  • image(eqn 4)

where aij is the matrix element of the ith row and jth column. As the elasticity values of all elements in a matrix sum to 1, these elasticity values can be directly compared among different matrices (e.g. for different populations) to investigate whether populations differ in which transitions are most important for population growth.

All nine (out of 16) transitions from a rosette class (S, M or L) to the same or another rosette class theoretically consist of both a survival and a reproduction component. The transition S–S, for instance, contained both the survival of small plants that stayed small till the next year and the contribution of small plants that flowered and produced new small plants the next year. To be able to study the importance of these separate vital rates (i.e. life-cycle components), we also calculated the elasticity values of all vital rates (i.e. the proportional response of λ to small, proportional perturbations of each vital rate; Caswell 2001; Franco & Silvertown 2004):

  • image(eqn 5)

in which inline image is the elasticity value of the kth vital rate, w. The elasticity values of the vital rates do not sum to 1 like the elasticity values of the matrix elements. When an increase in a vital rate week lowers λ, inline image may even be negative. To be able to compare the inline image of different populations we therefore rescaled the absolute values of all inline image in each matrix to sum to 1 (Franco & Silvertown 2004).

scenario analysis

We analysed the effect of three different factors (floral herbivory by insects, rosette damage by sheep and drought stress for seed germination) that may control the dynamics of native range C. nutans populations. We estimated these effects using the increase in population growth rate when one or more of the factors were omitted from the model. Sheppard et al. (1989) compared the survival of rosettes that were damaged by sheep with the survival of undamaged plants in these populations during the summer of 1986. Mortality of undamaged plants was about one-third lower than the mortality of damaged plants. Our first scenario was therefore to simulate the absence of sheep damage by reducing mortality (1 − σj) of S, M and L by one-third to investigate the impact of the damage by sheep on λ. Twenty-seven per cent of the observed plants had no signs of sheep damage. In the ‘no sheep damage’ scenario, the impact of sheep may therefore have been slightly overestimated as the scenario simulates the potential effect in the case when all plants are damaged.

The second factor we examined was seedling recruitment, which differed strongly between years (Sheppard et al. 1989). To investigate whether deviations from the best possible environmental conditions for recruitment limited the population growth, we fixed the recruitment rates (ɛ) at 0·2, which is approximately the observed maximum. This level was about twice the average of the observed recruitment rate in the sowed plots and similar to recruitment rates found in invasive populations (Woodburn & Sheppard 1996; Shea & Kelly 1998). In the model, this increased seed germination was compensated for by reducing both the proportion of seeds remaining (σ1) and entering (ν) the seed bank (as they now germinated immediately).

For the third factor, we simulated population dynamics with no predispersal seed losses to insect floral herbivory (i.e. ϕ = 1). Nodding thistle capitula are attacked by a range of specialist insects including tephritid flies and weevils. The fly Tephritis hyoscyami L. is the first of these, attacking the early bud stage. Then the receptacle weevil Rhinocyllus conicus Frölich (a specialist frequently occurring on European C. nutans plants; Zwölfer & Preiss 1983; Zwölfer & Harris 1984) and gall fly Urophora solstitialis L. attack the flower heads later on in the bud stage, utilizing the flow of resources in the receptacle of the developing flower head. Other receptacle weevils, Larinus jaceae F. and Larinus sturnus Schaller, then oviposit on or in the bracts of the developing flower heads, where the larvae feed on the receptacle. The seed losses as a result of each particular insect species were variable in our populations, but the sum of the seed losses caused by all insects that fed on the flower heads was stable over the sites and years (90·2% on average, n= 6, SE = 2·7; Sheppard, Cullen & Aeschlimann 1994).

Finally, we modelled four additional scenarios in which we combined two or all three of the above scenarios (no damage by sheep, best establishment and no seed losses to floral herbivory). For each scenario we again bootstrapped the underlying databases to estimate 95% confidence intervals.


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

natural population dynamics

In total, 1095 plants were monitored within the sampled quadrats. The average density was 9·1 plants m−2 (Table 1). The projected growth rates (λ) of all three populations were well below unity (0·566 on average) and their 95% confidence intervals overlapped (Table 2). The matrix element elasticities showed that in population 2 transitions to and from the L class contributed more to λ than in the other two populations (Fig. 2). Reproduction also contributed more to λ in population 2: the absolutes of the rescaled elasticity values of all vital rates involved in reproduction summed to 61% vs. 38% and 33% in populations 1 and 3, respectively. However, survival and growth, especially of plants in the S class, were more important in the latter two populations (Fig. 1). The negative vital rate elasticities for bolting in the L class reflected the fact that in these declining populations survival of the largest plants contributed more to λ than fecundity, because bolting plants died. Note, however, that although these negative elasticities for bolting meant that lowering bolting increased these low population growth rates, population sizes would never have grown without reproduction.

Table 2.  Population projection matrices for three French populations of Carduus nutans. The matrix classes are: SB, seed bank; S, small plants; M, medium-sized plants; L, large plants. The population growth rate, λ, and its bootstrapped 95% confidence interval are also given
Population 1, λ = 0·468 (0·361–0·584)
Population 2, λ = 0·627 (0·393–0·967)
Population 3, λ = 0·602 (0·533–0·820)

Figure 2. Year-to-year transitions (i.e. matrix elements) of the three studied populations. The elasticity values are given with each transition and are also indicated by the thickness of the transition arrows. Transitions with elasticity values smaller than 0·010 are not depicted. Matrix classes are: SB, seeds in the seed bank; S, small plants; M, medium-sized plants; L, large plants.

Download figure to PowerPoint

factors limiting population growth

The population growth rates were most sensitive to release from predispersal seed losses to floral herbivory by insects and far less sensitive to releases from sheep damage or below-maximum seedling establishment (Fig. 3). Release from insect herbivory on the seed heads rendered all significantly declining populations stable (λ ≅ 1). Only in population 2 did elimination of either of the other two controlling factors raise the projected population growth rate to near stability (the 95% confidence intervals included unity). The fact that maximizing seedling establishment increased λ slightly more than removing sheep damage in population 2, the opposite of the ranking in the other populations, showed that elasticity patterns predicted the response to larger manipulations rather well in this scenario analysis, even though elasticity is by definition a measure of the population response to small perturbations only. The combined effect of no sheep damage and the elimination of one of the other two factors could well be explained by the relative increases caused by eliminating those factors separately (Fig. 3; only 3% of the λ increase unaccounted for). However, the interactions were stronger in the combined no seed loss and best seedling establishment scenario and in the three-way elimination scenario where the increase in λ was underestimated by the product of the effects of the single factors by on average 20%, especially in population 2. The 95% confidence intervals increased with the increasing projected population growth rates. Even so, the populations were projected to grow significantly (the 95% confidence interval did not include unity) when either sheep damage or seedling establishment limitations were decreased in conjunction with the absence of floral herbivory by insects in populations 2 and 3, and in the three-way scenario in all populations.


Figure 3. Projected growth rates of three populations of Carduus nutans in France. In the different simulations we excluded (i) mortality as a result of sheep, (ii) low seedling establishment and (iii) predispersal seed losses as a result of insect floral herbivory, one at a time or in different combinations. Error bars denote bootstrapped 95% confidence intervals.

Download figure to PowerPoint


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

Specialist insect herbivores have had a very large impact on the observed native populations of C. nutans in southern Europe, as predicted by the enemy-release hypothesis (Keane & Crawley 2002). Dry growing conditions for germination and seedling establishment, as well as damage by sheep, limited the population growth to a lesser extent than the specialist insect floral herbivores, even with the extreme impact estimates used.

However, the impact of seed loss depended on the importance of reproduction for the growth rate of a particular population (Ehrlén 1996; Shea et al. 2005). In population 2, at the fertile site, fecundity had high elasticity values (as a result of higher flowering probabilities) and the absence of insects resulted in projected population growth rates (λ) twice as high as when insect herbivores that feed on flower heads were absent in the other populations (on less fertile sites), even though seed loss was more severe in the latter populations. When seed losses are excluded, saturation of the available microsites may start to limit the number of seedlings (Fröborg & Eriksson 2003). However, Kelly & McCallum (1992) found rosette survival in C. nutans only decreased as a result of density dependence at seedling densities of 800 m−1 or higher. Experimental studies on other plant species (including thistles) have shown that increased seed production following insecticide application indeed increased subsequent seedling and flowering plant densities (Louda 1982; Louda & Potvin 1995; Kelly & Dyer 2002; Maron, Combs & Louda 2002). Any overestimation of the increase in λ is therefore likely to be small in our study. The much-debated impact of floral herbivores on a species’ dynamics (Hoffmann & Moran 1998; Moran, Hoffman & Olckers 2004; van Klinken et al. 2004) is clearly evident for this non-clonal monocarpic perennial.

Fixing recruitment at the maximum observed level had less effect than the scenario with no specialist herbivory. This difference arose mainly because recruitment (ɛ) was only doubled, while in the ‘no insects’ scenario the proportion (ϕ) of potential seeds surviving floral herbivory was increased tenfold: a higher relative increase in one variable of the product (βπϕɛ) that forms the reproductive pathway can be expected to generate a bigger impact (Jongejans, Soons & de Kroon 2006). In C. nutans, ɛ and ϕ were not always multiplied, however, because only ϕ contributed to the transitions towards the seed bank (S–SB, M–SB and L–SB) while only ɛ contributed to the transitions from the seed bank to established plants (SB–S, SB–M and SB–L). However, the λ elasticity for any component (vital rate or element) is equal to that for any other component within the same loop (van Groenendael et al. 1994). Two vital rates that occur in exactly the same loops will therefore have the same elasticity values, even when they occur in different matrix elements, as long as the elements are constructed using these vital rates in a multiplicative and not in an additive fashion. As this was the case for ɛ and ϕ (equation 3), their elasticity values were identical (Fig. 1). Furthermore, when recruitment was doubled, seed bank formation and survival were lowered in the model at the same time. In declining populations, stasis (i.e. individuals surviving and staying in the same stage class) can be an important contributor to λ (Oostermeijer et al. 1996) and here the reduction of the seed bank stasis partly reduced the positive effect of more recruitment.

Damage to the rosettes by sheep had less impact on populations than the factors related to reproduction. Exclusion of damage by sheep resulted in a 33% increase in λ in the two more infertile sites with slower growing plants. This implies that optimal management will also depend on resource levels (Shea & Chesson 2002). However, sheep damage occurred within a grazing-driven community where grazing was also acting to maintain the populations of other pasture species. Competition with other plant species will strongly reduce C. nutans growth when grazing is halted (Popay & Medd 1995).

Seed losses were mainly the result of a community of specialist herbivore species that partitioned resources in the flower heads in a variety of ways, complementing each other's herbivore pressure in space and time, rather than because of one herbivore species in particular (Sheppard, Cullen & Aeschlimann 1994). This might suggest that releasing a community of seed head-feeding insects would be the best strategy for the biological control of C. nutans in the exotic range, as a single agent might not achieve consistent and sufficiently high levels of seed loss to suppress invasive populations. The release of multiple biological agents, however, has certain unpredictable risks. First, the agents may compete more severely in the new environment, producing a combined effect less than that expected from the separate effects of each agent on its own (Briese 1991a), either because of phenological differences or differences in the partitioning of seed head resources in the absence of their co-evolved natural enemies. Woodburn & Cullen (1996) found that R. conicus suppressed the damage potential of U. solstitialis following releases of these two agents in Australia, because photoperiodic differences caused U. solstitialis to become active earlier when nearly all accessible flower heads were already occupied by the weevil. Milbrath & Nechols (2004a,b) also found phenology-driven competition between Trichosirocalus horridus Panzer and R. conicus on C. nutans in the USA. Such competition can be especially inefficient for control efforts if the better competitor is not the one that has the largest impact on the population growth rate (Briese 1991a, 1991b; Shea 2004). Secondly, the impacts of the different flower head species may not always be high and may depend on the key vital rates in invasive populations. These would be expected to differ from those in the native range, for example depending on the relative importance of reproduction within the life cycle. Lastly, introducing many natural enemies increases the risk of non-target effects (Louda et al. 2003; Pearson & Callaway 2003; Coombs et al. 2004). Following the successful biological control of C. nutans by R. conicus in Canada (Harris 1984), a strategy of releasing a whole community of specialist flower head-feeding natural enemies was used against various monocarpic Centaurea species without first understanding the population dynamics of these weeds in the USA in the 1970–90s. This guild of natural enemies has contributed little to successful biological control (Coombs et al. 2004) and some indirect non-target impacts have emerged (Pearson & Callaway 2003).

Given that three of the insect herbivores (R. conicus, U. solstitialis and T. horridus) found on native C. nutans plants have already been introduced in the invaded range, and that different species have been shown to have different efficacy in different parts of the invaded range (Shea et al. 2005), it is perhaps best to augment the management tools that are already in place. The first priority is therefore to understand which weed population processes in the recipient environment are key. In our model the combination of seed losses and low seedling recruitment controlled population growth more strongly than would be expected from the separate impacts. This synergism shows that the factors that influence the target population are interdependent. Such interactions, in this case between specialist herbivores and seedling establishment opportunities, need to be considered when managing invasive species (Shea & Kelly 1998; Shea & Chesson 2002). Seedling establishment could be reduced by preventing overgrazing in pastures to ensure fewer gaps in the vegetation are available to seedlings, and by grazing pastures at times that are critical for the development of thistle seedlings (Bendall 1973). Integrated management that not only focuses on biocontrol agents but also integrates multiple factors will be more likely to control this invasive weed effectively in the invaded range (Shea, Thrall & Burdon 2000; Grigulis et al. 2001; Huwer et al. 2005; Shea, Sheppard & Woodburn 2006).

In a recent review, Hierro, Maron & Callaway (2005) discussed the importance of studying invasive species in their native as well as their invaded ranges. They stress that demographic studies allow us to understand what ecological factors may limit population densities and growth in the native range. Hinz & Schwarzlaender (2004) list 39 studies that compare species in both their invaded and native range. However, very few of these studies used mathematical models to investigate the population dynamics in the native range (Rees & Paynter 1997; Hinz & Schwarzlaender 2004). Matrix projection models and vital rate elasticity analyses are useful tools for predicting the impact of different controlling factors on population growth. Importantly, in its native range (this study) as well as in the invaded range (Shea et al. 2005) the populations of C. nutans varied considerably in what components of the life cycle contributed most to λ. Outcomes appear to be predictably dependent on the characteristics of the focus population: in some populations factors that reduce plant growth have a bigger impact, while insects that cause seed losses can control populations where fecundity is more important for the population growth rate. Further demographic studies across the native range would help to improve our models and deepen our understanding; however, our study shows how modelling the population dynamics of an alien invasive species in its native range can help to generate insights for pest management in the invaded range.


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

We thank Jim Cullen for instigating the field studies and Jean-Paul Aeschlimann, Jean-Louis Sagliocco, Agnes Valin and Janine Vitou for assistance in data collection, Jessica Metcalf for her help with the data analysis. And we are grateful to Zeynep Sezen, Emily Rauschert, Jasper van Ruijven, Cliff Moran and three anonymous referees for comments on earlier drafts of the manuscript. Part of this work was supported by NSF (DEB-0315860) and USDA-CSREES (Biology of Weedy and Invasive Plants) NRI grant no. 2002-35320-12289 to K. Shea and Australian Commonwealth Government support to A. W. Sheppard.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Bendall, G.M. (1973) The control of slender thistle Carduus pycnocephalus L. and C. tenuiflorus Curt. (Compositae) in pasture by grazing management. Australian Journal of Agricultural Research, 24, 831837.
  • Briese, D.T. (1991a) Demographic processes in the root-borer Agrilus hyperici (Coleoptera: Buprestidae): a biological control agent of St John's wort. Biocontrol Science and Technology, 1, 195206.
  • Briese, D.T. (1991b) Current status of Agrilus hyperici (Coleoptera; Buprestidae) released in Australia in 1940 for the control of St John's wort: lessons for insect introductions. Biocontrol Science and Technology, 1, 207215.
  • Buckley, Y.M., Rees, M., Paynter, Q. & Lonsdale, M. (2004) Modelling integrated weed management of an invasive shrub in tropical Australia. Journal of Applied Ecology, 41, 547560.
  • Buckley, Y.M., Rees, M., Sheppard, A.W. & Smyth, M.J. (2005) Stable coexistence of an invasive plant and biocontrol agent: a parameterized coupled plant–herbivore model. Journal of Applied Ecology, 42, 7079.
  • Burnside, O.C., Fenster, C.R., Evetts, L.L. & Mumm, R.F. (1981) Germination of exhumed weed seed in Nebraska. Weed Science, 29, 577586.
  • Carpenter, D. & Cappuccino, N. (2005) Herbivory, time since introduction and the invasiveness of exotic plants. Journal of Ecology, 93, 315321.
  • Caswell, H. (2001) Matrix Population Models Construction, Analysis, and Interpretation, 2nd edn. Sinauer Associates Inc., Sunderland, MA.
  • Colautti, R.I., Ricciardi, A., Grigorovich, I.A. & MacIsaac, H.J. (2004) Is invasion success explained by the enemy release hypothesis? Ecology Letters, 7, 721733.
  • Coombs, E.M., Clark, J.K., Piper, G.L. & Cofrancesco, A.F.J. (2004) Biological Control of Invasive Plants in the United States. Oregon State University Press, Corvallis, OR.
  • DeWalt, S.J., Denslow, J.S. & Ickes, K. (2004) Natural-enemy release facilitates habitat expansion of the invasive tropical shrub Clidemia hirta. Ecology, 85, 471483.
  • Efron, B. (1982) The Jackknife, the Bootstrap and Other Resampling Plans. Society for Industrial and Applied Mathematics, Philadelphia, PA.
  • Ehrlén, J. (1996) Spatiotemperal variation in predispersal seed predation intensity. Oecologia, 108, 708713.
  • Franco, M. & Silvertown, J. (2004) Comparative demography of plants based upon elasticities of vital rates. Ecology, 85, 531538.
  • Fröborg, H. & Eriksson, O. (2003) Predispersal seed predation and population dynamics in the perennial understorey herb Actaea spicata. Canadian Journal of Botany, 81, 10581069.
  • Grigulis, K., Sheppard, A.W., Ash, J.E. & Groves, R.H. (2001) The comparative demography of the pasture weed Echium plantagineum between its native and invaded ranges. Journal of Applied Ecology, 38, 281290.
  • Van Groenendael, J.M., De Kroon, H., Kalisz, S. & Tuljapurkar, S. (1994) Loop analysis: evaluating life history pathways in population projection matrices. Ecology, 75, 24102415.
  • Harris, P. (1984) Carduus nutans L., nodding thistle, and C. acanthoides L., plumeless thistle (Compositae). Biological Control Programs Against Insects and Weeds in Canada 1969–80 (eds J.S. Kelleher & M.A. Hulme), pp. 115126. Commonwealth Agriculture Bureau, Slough, UK.
  • Hierro, J.L., Maron, J.L. & Callaway, R.M. (2005) A biogeographical approach to plant invasions: the importance of studying exotics in their introduced and native range. Journal of Ecology, 93, 515.
  • Hinz, H.L. & Schwarzlaender, M. (2004) Comparing invasive plants from their native and exotic range: what can we learn for biological control? Weed Technology, 18, 15331541.
  • Hoffmann, J.H. & Moran, V.C. (1998) The population dynamics of an introduced tree, Sesbania punicea, in South Africa, in response to long-term damage caused by different combinations of three species of biological control agents. Oecologia, 114, 343348.
  • Huston, M.A. (2004) Management strategies for plant invasions: manipulating productivity, disturbance, and competition. Diversity and Distributions, 10, 167178.
  • Huwer, R.K., Briese, D.T., Dowling, P.M., Kemp, D.R., Lonsdale, W.M., Michalk, D.L., Neave, M.J., Sheppard, A.W. & Woodburn, T.L. (2005) Can an integrated management approach provide a basis for long-term prevention of weed dominance in Australian pasture systems? Weed Research, 45, 175192.
  • Jongejans, E., Soons, M.B. & De Kroon, H. (2006) Bottlenecks and spatiotemporal variation in the sexual reproduction pathway of perennial meadow plants. Basic and Applied Ecology, 7, 7181.
  • Kalisz, S. & McPeek, M.A. (1992) Demography of an age-structured annual: resampled projection matrices, elasticity analyses, and seed bank effects. Ecology, 73, 10821093.
  • Keane, R.M. & Crawley, M.J. (2002) Exotic plant invasions and the enemy release hypothesis. Trends in Ecology and Evolution, 17, 164170.
  • Kelly, C.A. & Dyer, R.J. (2002) Demographic consequences of inflorescence-feeding insects for Liatris cylindracea, an iteroparous perennial. Oecologia, 132, 350360.
  • Kelly, D. & McCallum, K. (1992) Evaluating the impact of Rhinocyllus conicus on Carduus nutans in New Zealand. Proceedings of the 8th International Symposium on Biological Control of Weeds (eds E.S. Delfosse & R.R. Scott), pp. 205211. DSIR/CSIRO, Lincoln University, Canterbury, New Zealand.
  • Kelly, D. & Popay, A.I. (1985) Pasture production lost to unsprayed thistles at two sites. New Zealand Weed and Pest Control Conference Proceedings, 38, 115118.
  • Van Klinken, R.D., Kriticos, D., Wilson, J. & Hoffman, J. (2004) Agents that reduce seed production: essential ingredient or fools folly? Proceedings of the XI International Symposium on Biological Control of Weeds, 27 April −2 May, 2003 (eds J.M. Cullen, D.T. Briese, D.J. Kriticos, W.M. Lonsdale, L. Morin & J.K. Scott), pp. 621622. CSIRO Entomology, Canberra, Australia.
  • De Kroon, H., Plaisier, A., Van Groenendael, J. & Caswell, H. (1986) Elasticity: the relative contribution of demographic parameters to population growth rate. Ecology, 67, 14271431.
  • Lee, J.M. & Hamrick, J.L. (1983) Demography of two natural populations of musk thistle (Carduus nutans). Journal of Ecology, 71, 923936.
  • Lonsdale, W.M., Farrell, G. & Wilson, C.G. (1995) Biological control of a tropical weed: a population model and experiment for Sida acuta. Journal of Applied Ecology, 32, 391399.
  • Louda, S.M. (1982) Limitation of the recruitment of the shrub Haplopappus squarrosus (Asteraceae) by flower-feeding and seed-feeding insects. Journal of Ecology, 70, 4353.
  • Louda, S.M. & Potvin, M.A. (1995) Effect of inflorescence-feeding insects on the demography and lifetime fitness of a native plant. Ecology, 76, 229245.
  • Louda, S.M., Pemberton, R.W., Johnson, M.T. & Follett, P.A. (2003) Nontarget effects: the Achilles’ heel of biological control? Retrospective analyses to reduce risk associated with biocontrol introductions. Annual Review of Entomology, 48, 365396.
  • Maron, J.L., Combs, J.K. & Louda, S.M. (2002) Convergent demographic effects of insect attack on related thistles in coastal vs. continental dunes. Ecology, 83, 33823392.
  • Medd, R.W. & Lovett, J.V. (1978) Biological studies of Carduus nutans (L.) ssp. nutans. I. Germination and light requirement of seedlings. Weed Research, 18, 363367.
  • Milbrath, L.R. & Nechols, J.R. (2004a) Indirect effect of early-season infestations of Trichosirocalus horridus on Rhinocyllus conicus (Coleoptera: Curculionidae). Biological Control, 30, 95109.
  • Milbrath, L.R. & Nechols, J.R. (2004b) Individual and combined effects of Trichosirocalus horridus and Rhinocyllus conicus (Coleoptera: Curculionidae) on musk thistle. Biological Control, 30, 418429.
  • Moran, V.C., Hoffman, J. & Olckers, T. (2004) Politics and ecology in the management of alien invasive woody trees: the pivotal role of biological control agents that diminish seed production. Proceedings of the XI International Symposium on Biological Control of Weeds, 27 April−2 May, 2003 (eds J.M. Cullen, D.T. Briese, D.J. Kriticos, W.M. Lonsdale, L. Morin & J.K. Scott), pp. 434439. CSIRO Entomology, Canberra, Australia.
  • Morris, W.F. & Doak, D.F. (2004) Buffering of life histories against environmental stochasticity: accounting for a spurious correlation between the variabilities of vital rates and their contributions to fitness. American Naturalist, 163, 579590.
  • Oostermeijer, J.G.B., Brugman, M.L., De Boer, E.R. & Den Nijs, H.C.M. (1996) Temporal and spatial variation in the demography of Gentiana pneumonanthe, a rare perennial herb. Journal of Ecology, 84, 153166.
  • Paynter, Q., Fowler, S.V., Memmott, J. & Sheppard, A.W. (1998) Factors affecting the establishment of Cytisus scoparius in southern France: implications for managing both native and exotic populations. Journal of Applied Ecology, 35, 582595.
  • Pearson, D.E. & Callaway, R.M. (2003) Indirect effects of host-specific biological control agents. Trends in Ecology and Evolution, 18, 456461.
  • Popay, A.I. & Medd, R.W. (1995) Carduus nutans L. ssp. nutans. The Biology of Australian Weeds (eds R.H. Groves, R.C.H. Shepherd & R.G. Richardson), Volume 1, pp. 2950. R.G. & F.J. Richardson, Melbourne, Australia.
  • Rees, M. & Paynter, Q. (1997) Biological control of Scotch broom: modelling the determinants of abundance and the potential impact of introduced insect herbivores. Journal of Applied Ecology, 34, 12031221.
  • Roberts, H.A. & Chancellor, R.J. (1979) Periodicity of seedling emergence and achene survival in some species of Carduus, Cirsium and Onopordon. Journal of Applied Ecology, 16, 641647.
  • Van Ruijven, J., De Deyn, G.B. & Berendse, F. (2003) Diversity reduces invasibility in experimental plant communities: the role of plant species. Ecology Letters, 6, 910918.
  • Shea, K. (2004) Models for improving the targeting and implementation of biological control of weeds. Weed Technology, 18, 15781581.
  • Shea, K. & Chesson, P. (2002) Community ecology theory as a framework for biological invasions. Trends in Ecology and Evolution, 17, 170176.
  • Shea, K. & Kelly, D. (1998) Estimating biocontrol agent impact with matrix models: Carduus nutans in New Zealand. Ecological Applications, 8, 824832.
  • Shea, K., Kelly, D., Sheppard, A.W. & Woodburn, T.L. (2005) Context-dependent biological control of an invasive thistle. Ecology, 86, 31743181.
  • Shea, K., Sheppard, A. & Woodburn, T. (2006) Seasonal life-history models for the integrated management of the invasive weed nodding thistle Carduus nutans in Australia. Journal of Applied Ecology, 43, 517526.
  • Shea, K., Thrall, P.H. & Burdon, J.J. (2000) An integrated approach to management in epidemiology and pest control. Ecology Letters, 3, 150158.
  • Sheppard, A.W., Aeschlimann, J.P., Sagliocco, J.L. & Vitou, J. (1995) Below-ground herbivory in Carduus nutans (Asteraceae) and the potential for biological control. Biocontrol Science and Technology, 5, 261270.
  • Sheppard, A.W., Cullen, J.M. & Aeschlimann, J.P. (1994) Predispersal seed predation on Carduus nutans (Asteraceae) in southern Europe. Acta Oecologica, 15, 529541.
  • Sheppard, A.W., Cullen, J.M., Aeschlimann, J.P., Sagliocco, J.L. & Vitou, J. (1989) The importance of insect herbivores relative to other limiting factors on weed population dynamics: a case study of Carduus nutans. Proceedings of the VII International Symposium on Biological Control of Weeds, 6–11 March 1988, Rome, Italy (ed. E.S. Delfosse), pp. 211–219.
  • Wardle, D.A., Nicholson, K.S. & Rahman, A. (1992) Influence of pasture grass and legume swards on seedling emergence and growth of Carduus nutans L. and Cirsium vulgare L. Weed Research, 32, 119128.
  • Woodburn, T.L. (1993) Host-specificity testing, release and establishment of Urophora solstitialis (L.) (Diptera, Tephritidae), a potential biological-control agent for Carduus nutans L., in Australia. Biocontrol Science and Technology, 3, 419426.
  • Woodburn, T.L. (1997) Establishment in Australia of Trichosirocalus horridus a biological control agent for Carduus nutans, and preliminary assessment of its impact on plant growth and reproductive potential. Biocontrol Science and Technology, 7, 645656.
  • Woodburn, T.L. & Cullen, J.M. (1996) Impact of Rhinocyllus conicus and Urophora solstitialis on achene set in Carduus nutans in Australia. Proceedings of the International Compositae Conference, Kew Gardens, UK (ed. N.J. Hind), Volume 2, pp. 307319. Royal Botanical Gardens, Kew.
  • Woodburn, T.L. & Sheppard, A.W. (1996) The demography of Carduus nutans as a native and an alien weed. Plant Protection Quarterly, 11 (Supplement 2), 236238.
  • Zwölfer, H. & Harris, P. (1984) Biology and host specificity of Rhinocyllus conicus (Froel.) (Col., Curculionidae), a successful agent for biocontrol of the thistle, Carduus nutans L. Journal of Applied Entomology, 97, 3662.
  • Zwölfer, H. & Preiss, M. (1983) Host selection and oviposition behavior in west-European ecotypes of Rhinocyllus conicus Froel. (Col., Curculionidae). Journal of Applied Entomology, 95, 113122.