Canopy cover and basal area are 2 common measures of tree cover used in forest wildlife habitat models and resource selection studies. When choosing between these 2 measures, it is important to recognize that they may differentially estimate relative cover of coexisting tree species due to differences in bole diameter distributions, crown overlap, and crown widths as a function of bole diameter. I found moderate agreement (multivariate p = 0.62, P < 0.0001) between lodgepole pine (Pinus contorta), Engelmann spruce (Picea engelmannii), and subalpine fir (Abies lasiocarpa) composition estimated by relative canopy cover and relative basal area in young to old-growth stands (n = 31) in subalpine forest of northcentral Colorado. However, differences between stand compositions estimated by relative canopy cover and relative basal area (average = 17%, range = 1-44%) were predictable based on multiple regression models of canopy cover as a function of basal area and trees/ha. I attributed most (11 of 14) above average deviations (>17%) to differences in diameter distributions among species in mature to old-growth (>200 yr) stands. Low densities of large diameter lodgepole pine or Engelmann spruce were a greater proportion of the basal area than of the canopy cover, while the converse occurred for higher densities of small and large diameter subalpine fir. Only 3 of 14 deviations >17% were attributed to species differences in crown overlap and crown width as a function of bole diameter, which occurred for young (22 yr) and intermediate-aged (48-80 yr) stands. I recommend use of basal area rather than canopy cover to estimate tree cover when it is desirable to emphasize large, uncommon trees that are resources used by the wildlife species of interest or that are indicators of important forest disturbance and successional conditions.