Weed evolution and community structure - Ecological Management of Agricultural Weeds - page 459

address the transformation of wild species into weeds (but see Jordan, 1989 a ,
1989 b discussed in section “Ecotype formation in weeds”below).
Baker (1991) described a series of events that indicate the sort of processes
that may be involved in the creation of new weeds. He noted that in the
gardens around Berkeley, California occurs a weedy, tetraploid race of Viola
alba , a species that is normally diploid with a chromosome number of 2 n 20.

He observed in his own garden a particularly vigorous cytological mutant
with 34 chromosomes.Although the mutant was sterile,it spread aggressively
by runners and perhaps could have become a significant weed had it not died
out during the drought of 1977-78.Alarge number of nonweedy and ruderal
species “test” genotypes in fields and gardens around the world each year.
Though most, as with Baker's Viola , are ultimately unsuccessful, the few that
do persist and thrive contribute to the weed problems of succeeding centuries.
Which species are most likely to evolve an agrestal habit is difficult to predict.
Presumably, these species must have some agrestal characteristics to begin
with (see Chapter 2), but, for example, are populations that spin off new weed
races most likely to be sparse or dense, continuous or patchy, outbreeding or
inbreeding?

Weedy races of crop species
A second avenue for the development of weeds is via the evolution of
weedy races of crop species. Such weeds can be extremely troublesome; by
mimicking key crop attributes,such as herbicide tolerance and seedling color-
ation, they prevent the use of selective control measures (Barrett, 1983). The
genetic changes that allow development of a weed race from a crop may come
about through mutation or through acquisition of characters from conspe-
cifics and congeners.
For grains, often all that is required to initiate a potential weed race is a
mutation that causes the inflorescence to shatter.Although additional charac-
teristics, such as nonsynchronous germination, early seed maturation, or
increased seed dormancy may be required to make the new form fully effective
as a weed (Baker 1965,de Wet 1975),the development of a shattering form ini-
tiates the process of evolution to weediness. Baum (1969) and Scholz (1986)
provided evidence for the mutant origin of shattering races of Avena sativa and
Hordeum vulgare . Harlan (1975) noted that although some of the African weed
sorghums disarticulate via formation of an abscission layer like the wild pro-
genitor, others disarticulate by breakage of the rachis.Domesticated sorghum
has lost the abscission layer but has a strong rachis and thus retains seeds on
the plant until harvest. The rachis-shattering weed races likely originated
from domesticated sorghum, since rachis breakage would be pointless prior