G. graminis var.avenae(Fig. 12.17). These spores would
not have sufficient nutrient reserves to infect if they
had to compete with the normal resident microbial ant-
agonists. But these antagonists are pH-sensitive, and
have been largely eliminated by the very low turf pH.
Chemical fumigants and fungicides (case 2) act in a
similar way, because they progressively reduce the
population of resident antagonists. And, sand beds
(case 3) have a naturally low resident population of
antagonists, because of their low organic matter con-
tent. The disease patches then spread rapidly for 2
or 3 years, killing the fine-leaved turf grasses and
enabling the coarser grass species and weeds to
invade. Then the patches stop spreading, but the
quality of the playing surface is seriously damaged.
Two different types of natural antagonist have been
implicated in the control of take-all patch disease.
First, antibiotic-producing fluorescent pseudomonads,
which are known to suppress G. graminisin turf of near-
neutral pH. Second, the fungus Phialophora graminicola,
which is a nonpathogenic invader of the grass root
cortex (Fig. 12.18) and also is known to suppress G.
graminisin turf of near-neutral pH (Deacon 1973).
In summary, the key to preventing the take-all patch
disease of grass turf lies in the management of turf pH,
which must be maintained at a level that supports the
natural antagonists, Phialophora, and/or fluorescentpseudomonads. This example illustrates that competi-
tion for a resource (competitive niche exclusion) can
be a highly effective means of disease control.Frost damage, and postharvest decay of
fruitLeaf and fruit surfaces support a characteristic popula-
tion of epiphytic bacteria and yeasts. Among these
are strains of Pseudomonas syringaethat have natural
surface proteins with ice-nucleation activity. The pres-
ence of these bacteria on plant surfaces promotes the
formation of ice crystals when the air temperature
falls even slightly below 0°, resulting in frost damage.
If these bacteria are absent then the temperature can
cool to −4° or lower before icing occurs, and this
difference can be important for frost-sensitive plants.
Lindow and his co-workers (see Lindow 1992) showed
that mutant strains (ice−) of P. syringaelacking the
surface protein can be pre-inoculated onto leaves to
exclude the wild-type (ice+) strains and give significant
protection against frost injury – another example of
pre-emptive competitive niche exclusion.
Other strains of P. syringae, and strains of the
yeast Candida oleophila, have been exploited to protect
against postharvest damage to fruit. Even slight252 CHAPTER 12
Fig. 12.17(a,b) The neck of a perithecium of G. graminisvar avenae, with long (c. 80μm) ascospores that are released
from the apical ostiole and can infect turf grasses that are not protected by natural antagonists.
(a) (b)