Weed evolution and community structure - Ecological Management of Agricultural Weeds - page 477

information on how changes in a character affect exposure to the mortality
factor would be needed to predict whether a population is likely to evolve
resistance to a given practice. Gathering such data would require consider-
able effort. However, since ecological management practices are primarily
useful within an integrated program, the potential of a weed to adapt to a
single tactic may be irrelevant.A reasonable first step in assessing the ability
of weeds to adapt to integrated ecological management would be to use
common garden or reciprocal planting techniques to compare the ecological
characteristics of weed populations from several pairs of adjacent conven-
tional and organic farms. This would at least reveal whether significant
divergence in response to cropping practices appears to occur over time
scales relevant to management.
Due to doubts as to the importance of weed adaptation and the dearth of
research on management of adaptation to control factors other than herbi-
cides, the following sections provide few prescriptions. Instead, they are
intended to indicate factors that may impact management of weed adaptation
should such management prove worthwhile.
Management of weed adaptation: basic concepts
Jordan & Jannink (1997) identified three general levels of manage-
ment of weed evolution.First,efforts may be directed toward limiting adapta-
tion to specific weed control tactics. Most of the literature on management of
weed evolution deals with this subject. Second, attempts can be made to
prevent adaptation to systems of weed management. Such adaptation may
involve evolution of several specific traits that provide simultaneous resis-
tance to each of the component weed management tactics in the system. In
this case, the management goal is to create selective regimes that make the co-
occurrence of all the necessary traits in a single genotype unlikely.
Alternatively, highly plastic, general-purpose genotypes may allow weeds to
be successful within integrated weed management systems. Avoiding selec-
tion for plasticity in diversified cropping systems may be difficult. Third,
weeds exist as metapopulations,systems of semi-isolated subpopulations that
can facilitate rapid evolutionary change (Wade & Goodnight, 1991; Gould,
1993; Hastings & Harrison, 1994). Controlling evolutionary processes driven
by metapopulation structure is likely to involve managing gene flow between
subpopulations. For example, work by Paulson & Gould reported in Gould
(1993) showed that increase of adaptive recessive genes could not occur if gene
flow between subpopulations exceeded a threshold value. In other situations,
gene flow may facilitate adaptive evolution of weeds (see section “Controlling
the spread of new weeds” below). The evolutionary consequences of