In central interior British Columbia, extensive cut blocks to recover timber killed by the Mountain Pine Beetle (Dendroctonus ponderosae) could impact negatively on Mule Deer (Odocoileus hemionus) late-winter habitat. This study assessed the possibility of predicting the distribution of potential late-winter habitat for Mule Deer with the BC Vegetation Resources Inventory (VRI) dataset used to produce forestry maps. On the basis of literature review and roadside inventories in December 2004, I predicted that high-quality Mule Deer late-winter habitat would correspond to mature and old conifer-dominated stands with ≤ 20% deciduous species, a canopy closure ≥ 45%, tree heights ≥ 23 m, tree diameter at breast height ≥ 24 cm, and basal area ≥ 45 m2/ha, and would be located on < 60% slopes on south, southeast, southwest or west aspects, or on flat ground. I allocated weight values to these criteria to classify map polygons into high-, medium- and low-quality polygons, and produce predictive maps of late-winter habitat use by Mule Deer. I tested my predictive habitat rating by snowtracking along 18 km of transects in February-March 2006, and 15.6 km of transects in February 2007. I recorded 31 and 12 Mule Deer tracks in 2006 and 2007, respectively, all in high-quality polygons. The observed frequency of tracks per polygon type was significantly (P < 0.001) different from expected. All tracks were in mature and old conifer-dominated stands including 10-60% Lodgepole Pine (Pinus contorta) and 10-20% Trembling Aspen (Populus tremuloides). This study showed that it was possible to predict the distribution of potential late-winter habitat for Mule Deer using a series of habitat criteria and the VRI dataset. The extensive harvesting of Lodgepole Pine in mixed coniferous stands will undoubtedly have a negative impact on Mule Deer late-winter habitat quality and quantity. The rating of habitat types developed in this study should be used in forest management plans to determine sites that should be protected from logging.