Weed evolution and community structure - Ecological Management of Agricultural Weeds - page 481

has greater risk of generating management-resistant races of species with seed
banks. For example, Gressel (1991) noted that Senecio vulgaris first evolved tri-
azine resistance in orchards, nurseries, and roadsides rather than in maize
fields. Few S. vulgaris seeds persist longer than one year on the soil surface
(Popay & Roberts, 1970), whereas many remain viable for several years when
incorporated into the soil (Watson, Mortimer & Putwain, 1987).
A seed bank also biases selection toward traits that are favored during years
with high seed production (Templeton & Levin,1979).Conversely,selection for
traits that are favored during years when few seeds are produced has little
effect on the population, because seeds carrying those traits are few relative to
those from years of high seed production. Consequently, when a weed with a
persistent seed bank first begins to show resistance to a control tactic, use of
additional measures that reduce seed production will slow the rate of adapta-
tion. Analogously with seed production, selection during years when emer-
gence from the seed bank is great is more effective in changing the genetic
makeup of a population than selection in years when emergence rate is low.
Essentially,a seed bank provides a mechanism whereby evolution is partly gov-
erned by the absolute quality of the environment as measured by seed produc-
tion and percentage emergence. One consequence is that adaptation to control
tactics that occasionally fail is likely to be slow for weeds with seed banks.
Controlling the spread of new weeds
A key part of any general strategy for management of weed commu-
nities and the adaptation of weed populations is prevention of dispersal
between fields and regions, and rapid response to new forms once they are
present. This requires a mode of operation that goes beyond the usual pre-
scriptive approach to the management of weeds with herbicides and even
beyond the integrative approach advocated throughout this topic. Rather,
managing the spread of new weed species and biotypes in a region requires an
approach analogous to that used to control communicable diseases, with
emphasis on epidemiological analysis, education, and societal response
(Green, 1990, pp.254-85).
Preventing the spread of weeds between fields and regions is a critical and
much neglected aspect of weed science. Blocking movement of weeds clearly
reduces the rate of increase in the number of infested fields for an alien species
that is spreading within a region. Less obviously, it also reduces the total
number of fields infested at equilibrium by shifting the balance between colo-
nization and local extinction (see section “Species introduction and the
species richness of weed communities”above).