Weed evolution and community structure - Ecological Management of Agricultural Weeds - page 478

metapopulation structure in weed populations require exploration through
modeling, model systems, and case studies.
Discussion in the weed science literature of methods for managing the
adaptation of weeds to specific control measures has focused primarily on pre-
vention of herbicide resistance. Although some analogies may be made
between control of herbicide resistance and control of adaptation to ecological
management practices, such analogies are limited by differences in the mode
of inheritance. Resistance to herbicides and other anthropogenic toxins is
usually controlled by one to a few major genes (Macnair, 1991; Jasieniuk,
Brûlé-Babel & Morrison, 1996). First, because herbicides usually target a
single enzyme, a small but highly specific change in the genome is sufficient
to confer resistance. Second, mortality from an effective herbicide is high
(often 95% to 99%), and consequently the amount of phenotypic change
required in a single generation usually exceeds the available additive genetic
variation for tolerance in the population. Intermediate forms cannot survive,
and hence one to three genes with large effects usually control herbicide resis-
tance. However, the degree of herbicide tolerance to low doses or in naturally
tolerant species is usually controlled by many genes (Putwain & Collin, 1989).
In contrast to herbicide resistance, characters like growth rate, stature, seed
mass, seed longevity, and phenology of germination that might change
during adaptation to ecological management practices are quantitative char-
acters controlled by many genes.
Several strategies for slowing or preventing the evolution of herbicide
resistance have been suggested based on genetic models. These include (i)
slowing the rate of increase in the frequency of resistant genes by rotating use
of a given herbicide with other chemical and nonchemical controls (Gressel &
Segel, 1990; Gorddard, Pannel & Hertzler, 1996; Jasieniuk, Brûlé-Babel &
Morrison, 1996), (ii) decreasing the probability of resistant individuals in the
population by use of multiple herbicides simultaneously (Gressel & Segel,
1990), (iii) decreasing selection pressure for resistance by using rates that
allow some escapes (Maxwell, Roush & Radosevich, 1990), and (iv) leaving
untreated areas to provide flow of susceptible genes into the population
(Maxwell, Roush & Radosevich, 1990).
With the exception of rotating control methods, these tactics are largely
irrelevant to ecological weed management. Gene flow occurs over such short
distances in most weed species that the last approach would usually require
leaving unacceptable numbers of uncontrolled individuals (Jasieniuk, Brûlé-
Babel & Morrison, 1996; but see discussion of spatially complex cropping
systems in the next section). The second and third approaches apply specifi-
cally to adaptation via major genes. In particular, decreasing selection pres-