ARTIFICIAL VS. FIELD INOCULATION KITH FUSIFORM RUST 



333 



RESULTS 



ARTIFICIAL INOCULATION 



Variation among families was significant in both 1963 and 1964. In 

 both tests, the proportion of diseased offspring from one rust-free and 

 the three infected selections was greater than from the two remaining 

 rust-free selections (Fig. 1) . The combined analysis of variance con- 

 firmed the significance of the family variance, but both the year x 

 family interaction and differences between years were non-significant. 



FIELD INFECTION 



Stem infections appeared within 1 year of planting at both locations, 

 Since then, the number per plant has increased steadily and significant 

 differences among families were noted as early as 2 years after planting. 

 Although few changes in family ranking have occurred since, considerable 

 variability was encountered within families. For example, the coeffi- 

 cient of variation at Gulfport ranged from 108% during the second year 

 to 65% in the fifth year. By then, progenies of 18-40 and 18-62 had 

 more stem infections than those of any of the other four selections 

 (Table 1). In addition, progeny of 8-7 at Gulfport had fewer stem infec- 

 tions. Simple correlations between numbers of stem infections and other 

 infection indices were significant at the 0.01 level. 



Table 1. Mean numbers of fusiform rust infections per plant 

 among open-pollinated slash pine progenies in the field 



Location 



Family 



Number of infections per plant 



Stem 



Total 



Gulfport 

 year 5 



Bogalusa, 

 year 4 



18- 



62 



18- 



40 



18- 



-27 



11- 



- 6 



18- 



-41 



8- 



- 7 



18- 



-62 



18- 



40 



18- 



-27 



11- 



- 6 



8- 



- 7 



0.65 

 .60 

 .36 

 .32 

 .31 

 .02 



.42 

 .40 

 .20 

 .14 

 .04 



1.71 

 1.72 

 .87 

 .62 

 .78 

 .04 



1.46 

 1.90 

 1.19 



.52 

 .05 



Values connected by the same vertical line do not differ signifi- 

 % cantly at the 0.01 level according to Duncan's test. 



