Fig 1.
Map of northern Borneo showing mountains and potential distribution of Mountain Treeshrews based on its elevation range (areas above 800 m). Areas between 800 and 1600 m are highlighted in light gray, those between 1600 and 2400 m in dark gray and those over 2400 m in black.
Fig 2.
Elevational distribution of body size.
Distribution of body size, approximated by (A) weight (W, in g) for D1 individuals and (B) head-and-body length (HBL, in mm) for D2 individuals across the studied elevational gradient. Vegetation zonation is indicated with vertical dotted lines. (A) Dot relative shading corresponds to the year of collection. Weight (W) followed a significant U-shaped distribution along elevation (P = 0.0337), with no significant differences between sexes (P = 0.8292), but a significant trend to increase over the years (P<0.0001) (W model R2adj = 0.2498, F4,79 = 7.909). (B) Black dots represent females, gray dots represent males. Head-and-body length (HBL) was not significantly affected by any of the response variables studied (HBL model R2adj = -0.0326, F4,79 = 7.909).
Fig 3.
Elevational distribution of craniodental variation.
Distribution of skull measurements taken from D6 individuals across the studied elevational gradient. Vegetation zonation is indicated with vertical dotted lines. (A) Condyle-incisive length (CIL), (B) rostrum length (RL), (C) upper tooth-row length (UTL), and (D) zygomatic breadth (ZB) across elevation. The limits of the vegetation zones are marked with vertical dotted lines. Black dots: females; gray dots: males. Lines show the relationship of the dependent variable with elevation for both sexes (only one line in the plot), for females (black lines) and for males (gray lines). (A) CIL followed a significant U-shaped distribution along elevation (P = 0.0038). Females showed significantly smaller CIL values than males (P = 0.0096) (CIL model R2adj = 0.2363, F3,102 = 11.83). (B) RL significantly decreased linearly with elevation (P = 0.0148). No significant differences were found between sexes (RL model R2adj = 0.8127, F3,94 = 141.3). (C) UTL followed a significant U-shaped distribution along elevation (P = 0.0378). No significant differences were found between sexes, but differences among mountains were significantly supported (P = 0.0167) (UTL model R2adj = 0.818, F6,98 = 78.93). (D) ZB significantly increased linearly with elevation (P = 0.0004). Females showed significantly lower ZB values than males (P = 0.0079). ZB significantly increased over the years (P = 0.0177) (ZB model R2adj = 0.3857, F4,101 = 17.48). CIL had a significant effect over RL (P< 0.00001), UTL (P< 0.00001) and ZB (P< 0.00001).
Fig 4.
Elevational distribution of body shape.
Distribution of (A) tail length (TL) of D2 individuals, (B and D) hindfoot length (HFL) of D3 individuals, and (C) ear length (EL) of D3 individuals (in mm) over the elevational gradient studied. Vegetation zonation indicated with vertical dotted lines. Black dots: females; gray dots: males. Lines show the relationship of the dependent variable with elevation for females (black lines) and for males (gray lines). (A) Tail followed a U-shaped distribution over elevation (P = 0.0163). Females had significantly shorter tails than males (P = 0.0321) (TL model R2adj = 0.353, F4.46 = 7.821). (B and D) Although HFL globally significantly increased linearly over elevation (P = 0.0011), significant differences were observed between Mt. Kinabalu (B, N = 22) and Mt. Tambuyukon (D, N = 13, P = 0.0022). Females had significantly shorter hindfeet than males (HFL model R2adj = 0.5665, F4,30 = 12.11). (C) No significant trends were observed for EL. HBL had a significant effect only over HFL (P = 0.0018).
Fig 5.
Elevational distribution of dorsum hair length.
Distribution of dorsum hair length (in mm) at the rump (HRL) and scapula (HSL) of Mountain Treeshrews over an elevational gradient (in m) on Mt. Tambuyukon. The limits of the vegetation zones are marked with vertical dotted lines. Hair length (HRL, HSL) significantly increased with elevation (HRL model R2adj = 0.5618, F1,15 = 21.52, P = 0.0003; HSL model R2adj = 0.5033, F1,15 = 17.21, P = 0.0009).