Agriculture Reference
In-Depth Information
period, the
M. graminicola
population in a wheat field is a meta-population,
composed of very many sub-populations, each confined to a very small area. By
contrast, the ascospores of
M. graminicola
are wind-dispersed and are formed
throughout much of the growing season (Hunter
et al.
, 1999; Eriksen and Munk,
2003) Since ascospores can be dispersed over long distances, the selection pressures
which act on them are quite different from those acting on the small sub-populations
in which the ascospores are formed. Furthermore, the same selection pressures act
on ascospores over a large area, so those spores form a single, geographically
extensive population.
3.3.2 Sampling populations
(a) Single genotype and bulk isolates
Samples from pathogen populations may be collected as pure genotypes, such as
single colony isolates of powdery mildew fungi, single pustule isolates of rust fungi,
single pycnidium isolates of
Mycosphaerella
spp. and so on. Alternatively, bulk
isolates consist of many individuals, usually of many genotypes, and are usually
intended to be representative of the pathogen population on a plant, in a field or in
some larger area.
More precise information can be obtained from single colony isolates than from
bulk isolates, but if it is more important to collect information from many sampling
sites than to have very precise data about the population at any individual site, it
may be more cost-effective to use bulk isolates. However, if a bulk isolate has to be
maintained for several generations before being tested, there is a danger that some
genotypes will out-compete others, with the result that the bulk isolate, when it is
tested, is not actually representative of the population from which it was sampled.
(b) Sampling schemes
For population genetic analysis, the ideal sample is a random one, obtained from a
defined population. This presents two problems: how is the target population to be
defined and how can a random sample be collected from it.
Wind-dispersed spores of a few fungi can be sampled randomly and such a
sample can be treated as a random sample from the source population if it is
assumed that this population of spores will form the next generation of inoculum.
This is done by many laboratories for
B. graminis
, which is sampled by allowing
airborne conidiospores to infect susceptible trap plants. Such a sample should be
randomly drawn from the local conidial population from fields upwind of the
sampling site. However, a sample should not only be random, but also be
representative of the population in which one is interested. Hovmøller
et al.
(1995)
investigated stationary and mobile traps of
B. graminis
f.sp.
hordei
, the former being
plants of the susceptible variety Pallas placed more than 1 km from the nearest
barley field, the latter being leaves of Pallas placed in a spore trap mounted on the
roof of a car, driven around the area close to the stationary site. There was less
