Lists of figures and tables
Fig. 1 Distribution of sampling sites (a) and spatial patterns
(b and c) of trichome density in oriental oak across Eastern Asia. The
circles and triangles represent 44 in situ populations, triangles
represent 15 original sites of common garden populations, and the square
represents the location of the common garden. The picture on the right
represents the morphology of trichome in oriental oak. N, sample number;
Max, maximum; Min, minimum; SE, stand error; CV, coefficient of
variation.
Fig. 2 The piecewise structural equation model (SEM) for
testing the effects of environmental factors and leaf functional traits
on leaf trichome density. GeoPC1 and
GeoPC2 represent the first two principal components of
the 3 geographic variables, ClimPC1 and
ClimPC2 represent the first two principal components of
the 5 climatic variables, and eigenvector scores for two axes of
principal component analysis (PCA) are also given. The standardized
regression coefficient is shown for each significant path (P <
0.05). The conditional (c) and marginal (m) R2 are
provided for each exogenous variable
(R2c = both fixed and random effects,
and R2m = fixed effects). Black arrows
denote positive relationships, red arrows negatives ones. Blue arrow
denotes the correlation between SD and SS. LMA, leaf mass per area; SD,
stomatal density; SS, stomatal size; VD, vein density.
Fig. 3 Relationships between trichome density and environmental
factors for oriental oak (n = 44).
Fig. 4 Relationships between trichome density and leaf
functional traits for oriental oak.
Fig. 5 Difference in trichome density between the common garden
and in situ. The common garden is located in Shanghai (at 4 m above
sea-level, and 31°02′N and 121°26′E), where MAT and MAP is 15.5 °C and
1149.8 mm, respectively. The letters above bars represent a significant
difference at the P < 0.05 level, by Duncan’s test (n = 15).
Fig. 6 Relationships between trichome density and leaf
functional traits in the common garden (n =15) and in situ (n =15).