We evaluated two juniper removal treatments (Fall, Spring) to restore aspen woodlands in southeast Oregon, spanning a 15-year period. The Fall treatment involved cutting 1/3 of the juniper followed by a high severity broadcast burn one year later in October 2001. The Spring treatment involved cutting 2/3 of the juniper followed by a low severity broadcast burn 18 months later in April 2002. After 15 years, aspen density in the Spring treatment was about 1/3 of the Fall treatment, however, aspen cover did not differ from the Fall treatment. Because spring burning was less effective at removing juniper, leaving about 20% of the mature trees and 50% of the saplings, retreatment of conifers will be necessary to maintain the aspen community. If an objective is to maintain or increase native understories the Spring treatment was more effective than the Fall treatment for recovering the shrub layer.