We supposed that the leaf form of the physiologically favoured part of a tree would serve as a genetical indicator. Considering that the lower part of the crown of a SUGI (Cryptomeria japonica) tree is in such a condition, we made comparisons of the form of leaves obtained from this part among several races and among natural forests. From the results of those comparisons, we proved the existence of "ecocline" in the leaf angle of natural forests in the western part of Honshu.
We assumed that the leaf length of japanese two-leafed pines would be useful as the genetical indicator of this pine. Measuring leaves of all current year branchlet attached to one strong branch located on the upper part of an AKAMATSU (Pinus densiflora) tree, we found that leaves of the outer part of the branch were the longest. Measuring leaves on every part of this tree (in all the four directions and five layers of its crown), we found the longest on the sunny, upper two-fifth part of this crown.
Among 30 trees of five-year-old clone of AKAMATSU and between them and their mother tree, there was no significant difference in the leaf length. So, we concluded that the leaf length of the longest of a japanses two-leafed pine might be recognized as a genetical indicator of them.
Furthermore, we took leaves of several clones of japanese two-leafed pine, measured their length and examined the number and the location of resin ducts seen in their cross sections. Results are shown in Table 5 and 6. The relation between the leaf length and the degree of AKAMATSU-KUROMATSU hybridity (indicated tentatively by the ratio of the number of marginal resin ducts to the total number in the cross section) is not very significant.