THE PROPAGATION OF APPLE TREES. 



93 



deeply stained. In this way a fairly satisfactory criterion was established. 

 (See Plates III and IV.) Four to thirteen twigs of each variety were ex- 

 amined and five to ten measurements and counts of cambium cells made 

 on each twig. No differences were detected between shoots from nurs- 

 ery trees and from bearing trees. Table 12 gives the results of 

 measurements. 



Table 12. — Thickness of the Bark and Cambium. 



It appears from this table that there is a difference in the thickness 

 and number of cells in the cambium layer of the varieties examined, and 

 that this is correlated with the ability of the variety to form roots from 

 the scion. The only marked exception shown in the table is the Baldwin, 

 which, having the fewest cells and the thinnest cambium layer of all, roots 

 more freely than four of the other varieties studied. Further study of 

 this question, including other varieties and extending through the growing 

 season, should prove definitely whether we have here a significant reason 

 for the variation in root formation among different varieties. 



DISCUSSION OF THE RESULTS. 



As a major result of the work here reported two facts are brought out: 

 (1) varieties differ greatly in their readiness to form roots from the scion 

 when propagated by the nurse-root method; (2) there is also great varia- 

 tion within the variety in the number that form roots from the scion. 



Taking up first the varietal differences we find that a few varieties 

 root in all, or nearly all, cases, while only one variety of Pyrus mains — 

 Bethel — has failed entirely to yield trees rooted from the scion. Inas- 

 much as this variety was grown in rather small numbers and under con- 



