J. Read et al., Relationships between sclerophylly, leaf biomechanical properties and leafanatomy in some Australian heath and forest species, PL BIOSYST, 134(3), 2000, pp. 261-277
A previous study of 19 south-east Australian heath and forest species with
a range of leaf textures showed that they varied considerably in leaf biome
chanical properties. By using an index of sclerophylly derived from botanis
ts' rankings (botanists' sclerophylly index, BSI) we determined that leaves
considered by botanists to be sclerophyllous generally had both high stren
gth and work to fracture (particularly in punching and tearing tests), both
at the level of leaf and per unit leaf thickness. In the current study we
have shown that leaves from the same species also varied considerably in le
af specific mass 146-251 g m(-2)), neutral detergent fibre concentration (2
0-59% on a dry weight basis) and in leaf anatomy. Multiple regression indic
ated a very strong correlation between BSI and the first two components of
a principal components analysis (PCA) of leaf anatomy (R-2 = 0.91). In addi
tion, there was strong correlation between the first component of a PCA of
the mechanical properties (correlated with BSI) and the two axes derived fr
om anatomical characteristics (R-2 = 0.66). The anatomical properties contr
ibuting most to the cuticle (axis 1) and percentage fibre (neutral detergen
t fibre) and lower epidermis thickness (axis 2). However, whether these rel
ationships are causal, or reflect correlations with characteristics not mea
sured in this study, such as: vascularization and sclerification, is not cl
ear. At a finer scale, however, there is evidence that there are various wa
ys to be sclerophyllous, both in terms of anatomical and mechanical propert
ies. This is illustrated by comparison of two of the sclerophyllous species
, Eucalyptus baxteri and Banksia marginata.