An apple planting was established in 1996 comprised of two cultivars: `Ginger Gold' (GG) and `Crimson Gala' (CG) on Malling 9 NAKB T337 and Budagovsky 9 at the Horticulture Research Farm at Rock Springs, Pa. The trees were planted at a spacing of 1.5 × 3.7 m in a randomized complete-block design with 10 replications. The trees were trained to a vertical axe system with a single wire set at 2.8 m, to which the conduit was attached. Data collected included trunk cross-sectional area (TCA), tree yield, number of fruit, and number of rootsuckers. Calculated data included annual tree growth, tree efficiency, average fruit weight, and crop load. In most years, there were significant cultivar × rootstock interactions for some variables. At planting and for the first two growing seasons, GG/B.9 were significantly larger than GG/M.9 as measured by TCA. At planting, there were no differences in TCA for CG, but, by the end of 1996, M.9 trees were significantly larger and stayed this way for the rest of the study. The GG/M.9 trees did not have significantly larger TCA than those on B.9 until 2005. Trees on B.9 were 23% and 31% smaller in 2005 for GG and CG, respectively, for B.9 than on M.9. Flowering occurred first and in greater abundance for GG/B.9. At the end of the 10th growing season, there was no difference in number of fruit or total yield per tree within cultivars by rootstock. However, for both cultivars, efficiency was highest for trees on B.9. Rootsuckers were greatest for trees on B.9. Fruit weight, when adjusted with number of fruit/tree as a covariate, was different for GG in some years.