Articles
FIRE BLIGHT RESISTANCE IN CULTIVARS AND SELECTIONS OF PEAR AND THE CORRELATIONS BETWEEN MEASURES OF RESISTANCE
Succulent shoot tips on 5-year-old orchard and 3-month-old greenhouse trees were inoculated with about 0.1 ml bacterial suspension (5x 108 cells/ml) of a mixture of two E. amylovora cultures (Ea 266 and 273), using a 26-gauge hypodermic needle.
Any trees which escaped infection were re-inoculated.
For comparison, shoots on orchard trees were also inoculated with a mass inoculation clamp.
Disease reaction was measured by five methods: 1) mean length of shoot necrosis for all inoculated shoots; 2) mean percentage of shoot length showing necrosis; 3) percentage of total number of shoots showing necrosis; 4) mean length of shoot necrosis for only those shoots showing infection; and 5) mean percentage of shoot length showing necrosis for only the shoots showing infection.
In spite of a relatively large amount of variability in reaction within genotypes, analysis of variance indicated statistically significant differences in resistance/susceptibility for the cultivars and selections.
Percentage of infected shoots were generally highest for greenhouse needle inoculations with 22 of 38 genotypes being 100% infected, followed by orchard needle inoculation, with only 8 of the genotypes being 100% infected.
Inoculations utilizing the mass inoculation clamp were much less successful, with only one genotype totally infected.
The range of, and average coefficients of variability for the five measures were similar and not excessive.
Correlations between the five measures within both greenhouse and orchard needle inoculations were significant and in general all greater than 0.80. The percentage of shoots infected was correlated with the other measures, but the magnitude of the coefficients was lower (r=0.36 to r=0.79, depending on the measure). Only the correlation between greenhouse and orchard needle inoculations for percentage shoot necrosis (based on all shoots), r = 0.50, was significantly different from zero.
Correlations of the measures with mean epiphytotic fire blight score (USDA scale) were generally of low to moderate magnitude (r = 0.4 or lower).
These results indicate that care must be used in selecting or discarding selections from a breeding program based only upon greenhouse or orchard inoculations.
The extent of spread of 1-year old infections into older wood should also be noted to determine the best method of evaluating resistance of pear germplasm to fire blight.