Growth and survivorship of Machillus thunbergii saplings in mountain forest are likely to be affected by topographical factors, biotic interactions with neighbouring plants, and individual size. However, such effects are always species- and site-specific, and may influence how plant species diversity contributes to ecosystem productivity. This study aimed to examine how individual growth and survivorship of Machilus thunbergii saplings are affected by: (1) topographical factors, such as aspect and inclination of slope, (2) species richness and type of neighboring plants, and (3) individual plant size. The experiment was conducted in the framework of BEF-China, a manipulated subtropical forest site in China. A total of 265 plots of 25.82 m× 25.82 m were planted with 1, 2, 4, 8, 16 or 24 different tree species. Each plot contained 400 trees arranged in a rectangular pattern with 1.29 m distance between individuals. In 2010 we added 16 individuals of M. thunbergii to each plot. These 16 individuals were planted in two rows along the western edge of the plots, with each individual in the center of 4 adjacent trees. Height and basal diameter of 1,452 surviving Machilus saplings were measured in June 2011 and 2012. ANOVA and Duncan’s multiple comparison tests were used to analyze the effects of both topography and of neighbouring plants, and linear regression was used to test size-dependence of growth. We found that Machilus saplings on shady slopes grew faster and had higher survival rates than those on sunny slopes, while the height increment of Machilus on plots with a mild incline was greater than that on steep slopes. Richness of neighboring plant species had an insignificant effect on Machilus sapling growth and survival rate; while the type of neighbouring species had different effects on Machilus sapling growth, but no effect on survival rate. Deciduous broadleaved species had the greatest effect on growth, followed by a mixture of deciduous and evergreen broadleaved species, followed by evergreen broadleaved species and lastly by evergreen needle leaved species; Machilus sapling growth was positively size-dependent, i.e., larger saplings grew faster. We conclude that, modelling tree growth should simultaneously incorporate topographical factors, species-specific neighborhood interaction and size of individuals, thereby providing a more accurate prediction of forest productivity and development, as well as information that will aid the conservation of endangered species.