Little research has been conducted on magnesium (Mg) nutrition of taro [Colocasia esculenta(L.) Schott cv. ‘Bun Long']. In this study, we evaluated the effects of varying levels of Mg (0.0, 0.05, 0.1, 0.2, 0.4, and 0.8 mM) on taro plants grown hydroponically for 33 days. Magnesium treatment effects were evaluated for dry matter biomass, leaf area, and N, P, K, Ca, Mg, Na, Mn, Fe, Cu, Zn, and B concentrations of old and young leaves. Dry matter of leaves (young, old, and total), roots, corms, petioles, and total biomass were significantly higher in all plus‐Mg treatments than in the zero‐Mg treatment. These same biomass parameters were not different among treatments with Mg (0.05 to 0.8 mM). Leaf area (young, old, and total) did not differ significantly with varying levels of Mg. A quadratic model described the relationship between Mg levels in leaves and solution Mg (r2= 0.99). Young and old leaf Mg concentrations did not differ. Total leaf Mg concentration ranged from 0.07% to 0.42% for the lowest and highest Mg levels in solution, respectively. Leaf Mg effects on total leaf DM was best fit using segmented regression (r2= 0.95), with a corresponding critical leaf Mg concentration (95% of maximum predicted leaf DM) of 0.14%. No significant interactions were observed between Mg and other mineral nutrients. Critical leaf Mg concentration is based on the vegetative growth stage of taro and could be a key index for taro producers who emphasize vegetable leaf, rather than corm production.