Food distribution is one of important aspects in the study of primate behavioral ecology. Although food species and preference of Rhinopithecus roxellana have been investigated in great detail, few studies have focused on its food distribution. In the period between August 2006 and July 2008, we quantified diet composition via instantaneous scans at 30 min intervals for a group of R. roxellana in Qianjiaping (area: ca. 60 km2; altitudes: 1 500~2 663 m), Shennongjia, China (Fig. 1), and meanwhile investigated its food distribution characteristics and seasonally altitudinal variation across its home range via vegetation transects at 200 m altitudinal gradients. The transect locations were selected as follows: 3 mountains were randomly selected from those with peaks higher than each altitudinal gradient between 1 500 m and 2 300 m, and then 1 transect (120 m ? 10 m) was made at each altitudinal gradient approximately at the central position of the eastern slope and western slope of each mountain, respectively; there was only 1 mountain with the peak higher than the altitude of 2 500 m, 1 transect (120 m ? 10 m) was first made at the altitudinal gradient of 2 500 m approximately at the central position of the eastern slope and western slope of this mountain, respectively, and then 1 transect (60 m ? 10 m) was made at the same altitudinal gradient with a distance of at least 500 m from the first transect at the eastern slope and western slope of the same mountain, respectively; those mountains were not counted if they were covered by alpine meadows at each altitudinal gradient; there were 32 transects with the size of 120 m ? 10 m and 2 transects with the size of 60 m ? 10 m in total, covering the area of 3.96 hm2. Within each transect, the DBHs (diameter at breast heights) of all trees (woody plants with the DBH of ≥ 30 cm or the height of ≥ 5 m), and the length and width of the crowns (estimated shrub coverage: length ? width) for all shrubs (woody plants with the DBH of < 30 cm and the height of < 5 m, and all lianas) were recorded. The proportions of tree basal area and shrub coverage at each altitudinal gradient and across the whole area were calculated for plants occupying ≥ 5.0% of the diet (defined as IFPs: important food plants) in any given season. The coverage of fruticose lichens on the surface of all trees and shrubs within each transect was estimated at 5 levels: I = 0%, 0% < II < 5%, 5% ≤ III < 10%, 10% ≤ IV < 20%, V ≥ 20%. The proportions of tree and shrub individuals with each coverage level by fruticose lichens at each altitudinal gradient were calculated, and the relationships between these proportions and altitudinal gradients were analyzed using Spearman rank correlation tests. Results showed that the animals ate at least 15 species of plants (occupying 55.0% of the overall diet) and fruticose lichens (occupying 38.4% of the overall diet) (Table 1). A detailed description of diet had been reported in Liu et al. (2013). In general, the density of food distribution was low across the area. IFPs accounted for 4.2~11.5% of tree basal area and 1.3~13.9% of shrub coverage in any give season (Fig. 2~3). About 89% of trees and 81% of shrubs were not loaded by fruticose lichens, and further only 0.8% of trees and 6.5% of shrubs had the coverage level of V (Fig. 4~5). Both IFPs and fruticose lichens were unevenly distributed across altitudes and mainly found at the altitudes between 1 900~2 500 m (Fig. 2~5), which may determine the limits of ranging altitudes of the animals. Specifically, the proportions of tree basal area of IFPs were highest at the altitudes of 1 900~2 100 m in spring and summer and at the altitudes of 2 100~2 300 m in autumn and winter, respectively (Fig. 2). The proportions of shrub coverage of IFPs were highest at the altitude of 2 500 m in spring and at the altitude of 2 100 m in summer, and were extremely small at any altitude in autumn and winter (Fig. 3). The seasonally altitudinal variation of IFPs might cause the animals to seasonally forage at different altitudes. The proportions of lichen-loading trees increased with altitudes (r = 0.94, P < 0.01), ranging from 5.8% at the altitude of 1 500 m to 23.0% at the altitude of 2 500 m (Fig. 4). The proportions of lichen-loading shrubs also tended to increase with altitudes (r = 0.77, P = 0.07), ranging from 8.7% at the altitude of 1 500 m to 45.5% at the altitude of 2 100 m (Fig. 5).