The leafing intensity premium hypothesis tested across clades, growth forms and altitudes

1. Recent reports provide empirical evidence for a negative isometric scaling of (log)leaf number per unit shoot volume (i.e. leafing intensity) to (log)leaf size in woody plants. In this context, a theory of leaf size variation was anticipated, stating that natural selection should have favoured high leafing intensity, given that most tree and shrub species are small-leaved. Leaf size is thus proposed to be amere correlate of leafing intensity. This was coined the leafing intensity premium hypothesis. 2. I investigated the generality of the leaf size–number trade-off by testing its extension to a set of 60 small-stature species, growing along an altitude gradient. Also, I compiled data from 224 species and re-analysed their leaf size and leafing intensity. Finally, I explored the validity of the leafing intensity premium hypothesis to account for the leaf size–number trade-off, if the latter was sustained, and examined patterns of coordinated evolution of leaf size and leafing intensity across seed plant lineages. 3. Negative isometric scalingwas supported in all analyses, except when evergreens were considered in isolation. Frequency distribution of both leaf size and leafing intensity was severely right-skewed in the 224-species data set, which provided no evidence in support of the view that natural selection favoured species with high leafing intensity. 4. Node divergences in the phylogenetic tree showed a tendency for inverse coordinated evolution of leafing intensity and leaf size. However, interactions between environments and divergence patterns, and also node-wise peculiarities, were detected. 5. Synthesis. The leaf size–number trade-off is of general scope, although scaling exponents differ among evergreen and deciduous species. In addition, signals of coordinated evolution of both traits across the seedplant treewere detected. Still, the leafing intensity premiumhypothesis cannot be supported, since leafing intensity of present-day species is not particularly biased towards high scores.