Understanding invasions: the rise and fall of diffuse knapweed (Centaurea diffusa) in North America - Invasive Plants: Ecological and Agricultural Aspects - page 116

and produce seeds is often limited by nitrogen availability [5]. Since the allelo-
pathic agent of diffuse knapweed requires nitrogen, we suspect that a trade-off
between allocation of nitrogen for root exudates and allocation of nitrogen to
seed production occurs in this species. Our data suggest that knapweed in
grasslands containing sufficient competitors becomes seed limited once the
herbivore L. minutus becomes abundant. Thus, allocation to allelopathic mate-
rials in the face of significant top-down controls is not going to contribute to
the success of the species. We note that, in the context of either the enemy
release hypothesis or the evolution of increased competitive ability hypothesis
(see Inderjit et al., this volume), escape from herbivory and root pathogens
may have allowed Centaurea species to produce sufficient root exudates to
generate allelopathic chemicals that is described by the novel weapons hypoth-
esis. Hence, the existence of novel weapons, should it be proven to be a fairly
common trait of invaders could be a consequence of the absence of biological
controls in the invaded communities and, if this is the case, should be regard-
ed as a consequence of factors identified in the enemy release hypothesis.
Missing from our knowledge is the extent to which Centaurea diffusa and
soil biota of the Colorado Front Range are similar to other communities of C.
diffusa in other regions of North America. Certain populations of C. diffusa
may differ in their abundances of alleopathic agents, and certain populations
may have more soil pathogens and fewer soil symbionts in other regions (e.g.,
[22, 33]). If so, then our results on top-down controls in these areas might not
have generality. We strongly believe, however, that we have described the gen-
eral case. Top-down controls have been reported in Montana and British
Columbia [25, 26], areas with different climates, different soils, and, presum-
ably, very different soil biota. Second, C. diffusa was able to invade and dom-
inate in our region regardless of its specific chemistry or the specific compo-
sition of soil biota. Thus, for now at least, top-down controls in some and per-
haps most regions appear to negate whatever combination of invasive traits C.
diffusa has mustered. Not all of our study areas have responded as quickly as
the site shown in Figure 1, but sites that appear slower to increase in seed head
weevil abundance and decrease seed production have all started with much
larger densities and seed sources of knapweed.
Will the top-down effects we observed for C. diffusa be repeated for other
species of invasive Centaurea in North America? The demise of a dominant
species requires greatly reducing propagule pressure, plant survivorship, or sig-
nificant reductions in both of these variables. We suspect that short-lived spe-
cies that by necessity rely heavily on seeds as a mechanism for success will like-
ly be most susceptible to the top-down controls observed here. Longer-lived,
iteroparous plants such as C. maculosa may be less vulnerable to insects like the
seed head weevil for several reasons. First, annual seed production per plant
appears relatively low, implying that maximum seed head weevil numbers,
which are limited by the number of seed heads produced, will also be relative-
ly low. Densities of adult weevils on a per-plant basis therefore may never
match those numbers occurring on C. diffusa . This would diminish tissue dam-