Agriculture Reference
In-Depth Information
Figure 1. Flowering stem densities of diffuse knapweed at two monitoring sites where insects were
released in 1997. Seed head weevils were relatively uncommon until 2000. Each bar represents the
mean and one standard error of 30 samples taken from 1 m 2 quadrats at two sites over the eight year
period. Data for the 1997-2001 interval were reported in [24].
and cover of knapweed [24]. By 2004, the seed head weevils had expanded to
almost all remnant grasslands in the area, and two plots that had substantial
knapweed when first censused in 2001 had no flowering stems or rosettes of
this species in 2004.
Summer precipitation during the study interval was variable and typical of
semiarid grassland found in a continental climate. During the last five years
rainfall was average or above average in 2001, 2003, and 2004. The year 2000
was moderately dry and the autumn and winter of 2001 and all of 2002 were
extremely dry. Knapweed rosettes may have refrained from flowering in 2002
but persisted through the drought. Seeds germinated by substantial rains in
2003 produced a modest increase in knapweed in 2004, but at levels well
below those observed prior to 2001. Rosette densities of knapweed were about
50 plants m -2 in 1997 [24], but ranged from 1-3 plants m -2 at the release site
in 2004 (data not shown). Given the potential persistence of a seed bank for
this species [29] the decline in rosettes likely reflected both the reduction in
seed production as well as substantial mortality of seedlings.
A single Larinus weevil larva will consume all of the seeds found in a dif-
fuse knapweed seed head. Thus, the average seed production of knapweed
impacted by this insect is determined by the number of seed heads with wee-
vils (all contributing zero seeds) as well as the seeds produced in those seed
heads not containing weevils. The relationship is 'triangular', i.e., when wee-
vil abundance is high, seed production is uniformly low, but when weevil
Search WWH ::




Custom Search