When the incidence of infection was tabulated on the basis of tree height it 

 was found that comparable results were obtained; that is, a large percent of 

 the trees in the shortest height class was infected and a much smaller per- 

 cent of the trees in the tallest height class was infected. Since the trees 

 in each drainage were all in the same age class, and all dominant or co- 

 dominant, height may be regarded as a rough measure of vigor. The data thus 

 indicate that slower growing trees are less resistant to infection than 

 vigorous trees. The results of tabulating incidence of infection in each of 

 three site-moisture classes seemed to indicate that on wet sites (close to 

 the stream) trees are less subject to infection than on moist and dry sites. 

 Since the areas sampled are all steep, well-drained, and in a region of only 

 moderate rainfall, it is assumed that moisture is favorable to the tree, 

 increasing its vigor and thus its resistance to infection. The influence of 

 site upon the fungi on these areas is not known. Because Armillaria mellea 

 is known to live saprophytically on the roots of snags and stumps and to 

 send rhizomorphs radiately to nearby trees, it was thought that there might 

 be a relationship between the number of snags and stumps in the vicinity of 

 individual trees and the incidence of infection, but the results of analysis 

 indicate no significant relationship. 



3 » Relation of External Symptoms to Degree of Infection 



The most obvious external symptoms of unhealth which might result from 

 root infection are color of foliage and density of crown. Accordingly, the 

 data on degree of infection were tabulated to bring out any connection which 

 there might be between degree of root infection and development of these 

 symptoms. The results are given graphically in figure 2. 



Color of foliage is apparently strongly related to degree of root in- 

 fection (figure 2A) . Of the trees without disclosed root infection, 97 per- 

 cent had healthy green foliage and none had appreciable quantities of brown 

 foliage. At the other extreme, none of the trees with more than half of the 

 circumference of root collar killed had healthy green foliage, and the trees 

 in which the root collar had been completely killed (only two in this class) 

 had mostly brown foliage. 



Degree of root infection is also fairly well reflected in density of 

 crown (figure 2B). Of the trees without disclosed root infection, 82 per- 

 cent had dense to medium crowns and both of the trees in which the whole of 

 the root collar was killed had ragged crowns. However, the fact that 18 

 percent of the trees without disclosed infection had open or ragged crowns 

 indicates that density of crown is not a completely reliable indication of 

 root infection. 



4 » Value of Symptoms in Extensive Tree Examinations 



Having recorded certain external and internal symptoms of root infec- 

 tion, it seemed desirable to determine how far the readily determinable 

 symptoms could be relied upon to furnish an indication of the degree of root 

 infection. In conformity with results above, color of foliage was found to 

 furnish a moderately reliable measure of degree of root infection, although 

 it was not a dependable criterion of absolute infection. Density of crown 



