An experiment was conducted to determine coefficients of standardized total tract digestibility (CSTTD) of P in canola meal (CM) fed to gestating and lactating sows without or with a multi-enzyme complex (MC). Eight sows cannulated on day 40 of gestation were assigned at random to the 4 dietary treatments in a replicated 4 × 4 Latin square design. The 4 diets included 2 cornstarch-based diets with 313 g/kg solvent-extracted CM as the only source of P, without or with MC, a casein-cornstarch diet to determine ileal endogenous amino acid losses (amino acid digestibility data reported elsewhere), and a P-free diet to determine the endogenous P losses. All diets contained 3 g/kg titanium dioxide as an indigestible marker. Gestating sows were fed 3.0 kg/day, whereas, during lactation, sows had ad libitum access to diets. Samples were collected over 3 phases; mid-gestation, late-gestation, and lactation. Each phase had 4 experimental periods lasting 8 days. In each period, after 5-day acclimation to the experimental diets, fecal samples were collected by grab sampling via rectal palpation on day 6. Results indicated no differences in total tract digestibility of nutrients between mid-gestation, late-gestation, and lactation. Enzyme supplementation improved (P < 0.05) the apparent total tract digestibility of dry matter and P, and CSTTD of P during lactation. The CSTTD of P in CM was 0.445 (without MC) and 0.504 (with MC) in mid-gestation, 0.455 (without MC) and 0.492 (with MC) in late-gestation, and 0.442 (without MC) and 0.489 (with MC) in lactation. In conclusion, the average CSTTD for P in CM fed to sows was 0.447 and MC significantly improved the CSTTD of P in CM during lactation.