The objective of this study was to determine if the growth of sweet orange (Citrus sinensis (L.) Osbeck cv. ‘Valencia’) nonembryogenic callus could be regulated and controlled via the mineral nutrient components of the medium. The 14 salts comprising Murashige and Skoog (MS) basal medium were subdivided into five component groups. These five groups constituted the independent factors in the design. A five-dimensional hypervolume constituted the experimental design space. Design points were selected algorithmically by D-optimality criteria to sample of the design space. Growth of the callus at each design point was measured as % increase of fresh weight at 14 d. An analysis of variance was conducted and a response surface polynomial model generated. Model validation was conducted by mining the polynomial for design points to two regions—“MS-like” growth and MS 25% growth and comparing callus growth to predicted growth. Five of the eight selected MS-like points and three of the six MS 25% growth points validated, indicating regions within the design space where growth was equivalent to MS, but the salt combinations were substantially different from MS, and a smaller region where growth exceeded MS by greater than 25%. NH4NO3 and Fe were identified as important factors affecting callus growth. A second experiment was conducted where NH4NO3 and Fe were varied, thus creating a two-dimensional slice through the region of greatest callus growth and provided increased resolution of the response.