Abstract
Regeneration of Fagus crenata was examined in the experimental forests managed by a shelterwood system at the Kurosawajiri Research Site, Iwate Prefecture. In a stand where clearing of the forest floor was carried out after shelterwood logging (6 residual trees per ha) in 1948, the secondary forest dominated by F. crenata had been formed. In a stand without clearing of the forest floor after shelterwood logging (13 residual trees per ha) in 1969, the resulting secondary forest was dominated by Prunus grayana and Magnolia hypoleuca, and regenerating F. crenata trees were not frequent. In a stand with clearing of the forest floor after shelterwood logging in 1969, many regenerating F. crenata trees were found, representing the L-shaped size structure, but their distribution was limited to under and around the residual conspecific canopy trees and they seldom occupied the upper layer. These results demonstrate that stands in which F. crenata seedlings had been established owing to clearing of the forest floor do not always turn into F. crenata-dominated stands. In addition to the timing between the practices and seed production fluctuation, differences in the abundance of F. crenata seedlings before cutting may be an important factor causing the difference in the regeneration success of F. crenata between the two stands in which the forest floor treatment was carried out.
