Sixty logs of 17-year-old thinned, unpruned plantation-grown Eucalyptus dunnii representing 10 trees from each of 2 provenances were obtained from a field-trial in central Guangxi Province, China, and processed in a nearby sawmill to produce dried appearance-grade sawn wood suitable for manufacture of short-length flooring. This study's objectives included determining: interrelationships between log end-splitting and growth stress measured on standing trees and important wood behaviour during processing; relationships between log external features and sawn product recovery; the potential for improvement in processing and/or product quality with the application of a steam reconditioning treatment; and differences in product value and processing characteristics between two provenances of E. dunnii.Boards were air-dried and kiln-dried in two matched batches and one batch was subsequently given a steam reconditioning treatment. For prediction of processing characteristics, the measured displacement on standing tree stems associated with growth stress release was moderately useful for prediction of log end-splitting but not significant for any other characteristic tested, and log end-splitting was not a significant predictor of recovery or product value. Log diameter measures and sweep were the only significant predictors of recovery. Diameter was sufficiently large in some logs to meet D-grade requirements according to Australian (VicForests) log grading standards. The estimated mill door values of sawn product obtained per log cubic metre for D-grade and residual-grade logs (below VicForests sawlog specifications) were RMB 1045 and RMB 887, respectively (approximately AUD 190 and AUD 161, respectively). Steam reconditioning treatment significantly reduced cupping and increased width and thickness dimensions but the presence of severe non-drying-related defect limited the direct benefit of the treatment. There was no significant difference between provenances in product recovery, product value or log end-split severity.