Experiments were conducted to determine the relative bioavailability (RBV) of the calcium salt of the hydroxy analog of DL-methionine (MHA-Ca, 84%) to DL-methionine (DL-Met, 99%) as Met sources fed to pigs. In Exp. 1, 42 crossbred barrows (initial BW of 15.0 ± 0.7 kg) were allotted to 7 treatments in a N-balance study. The basal diet (BD) was formulated to contain 15.4% CP and 0.22% Met (70% of requirement). Diets included 1) BD, 2) BD + 0.025% DL-Met, 3) BD + 0.050% DL-Met, 4) BD + 0.075% DL-Met, 5) BD + 0.038% MHA-Ca, 6) BD + 0.077% MHA-Ca, and 7) BD + 0.115% MHA-Ca. An increase in dietary inclusion rates of both Met sources linearly increased (P < 0.01) N retained (g/d) and N retention (% of intake). Using linear slope-ratio regression, the RBV value of MHA-Ca to DL-Met for N retained (g/d) was 63.0% on a product-to-product basis (75.0% on an equimolar basis). In Exp. 2, 40 crossbred barrows (initial BW of 15.5 ± 1.5 kg) were allotted to 5 treatments in another N-balance study. The BD was formulated to contain 17.0% CP and 0.22% Met (70% of requirement). Diets included 1) BD, 2) BD + 0.030% DL-Met, 3) BD + 0.060% DL-Met, 4) BD + 0.046% MHA-Ca, and 5) BD + 0.092% MHA-Ca. Increasing levels of DL-Met or MHA-Ca increased N retained (g/d) and N retention (% of intake) linearly (P < 0.001) and quadratically (P < 0.05). Using linear slope-ratio regression, a product-to-product RBV value of MHA-Ca to DL-Met was 68.4% (81.4% on an equimolar basis) for N retained (g/d). In Exp. 3, 276 pigs (12 barrow and 11 gilt replicates; initial BW of 7.09 ± 1.1 kg) were used in 3 diet preference studies. Pigs were randomly allotted to 1 of 3 treatment comparisons of feed choice: 1) BD (0.23% Met) or BD + 0.07% DL-Met; 2) BD or BD + 0.0825% MHA-Ca, and 3) BD + 0.07% DL-Met or BD + 0.0825% MHA-Ca. Pigs consumed a higher percentage (55 vs. 45%; P = 0.008) of their total feed intake from the diet supplemented with 0.07% DL-Met in Comparison 1, but a lower percentage (45 vs. 55%; P = 0.003) of their total feed intake from the diet supplemented with 0.0825% MHA-Ca in Comparison 2. There was no diet preference for DL-Met or MHA-Ca in Comparison 3. The observed Met source preference differences occurred in the barrow replicates but not in the gilt replicates. These results demonstrated the mean RBV of MHA-Ca to DL-Met of 65.7% on a product-to-product (wt/wt) basis or 78.2% on an equimolar basis and that a preference for Met sources was observed in barrows but not in gilts.