J. Suomela et al., VARIATION AMONG AND WITHIN MOUNTAIN BIRCH TREES IN FOLIAGE PHENOLS, CARBOHYDRATES, AND AMINO-ACIDS, AND IN GROWTH OF EPIRRITA-AUTUMNATA LARVAE, Journal of chemical ecology, 21(10), 1995, pp. 1421-1446
Leaf quality of the mountain birch (Betula pubescens ssp. tortuosa) fo
r herbivores was studied at several hierarchical levels: among trees,
among ramets within trees, among branches within ramets, and among sho
rt shoots within branches. The experimental units at each level were c
hosen randomly. The indices of leaf quality were the growth rate of th
e larvae of a geometrid, Epirrita autumnata, and certain biochemical t
raits of the leaves (total phenolics and individual phenolic compounds
, total carbohydrates and individual sugars, free and protein-bound am
ino acids). We also discuss relationships between larval growth rate a
nd biochemical foliage traits. Larval growth rates during two successi
ve years correlated positively at the level of tree, the ramet, and th
e branch, indicating that the relationships in leaf quality remained c
onstant between seasons both among and within trees. The distribution
of variation at different hierarchical levels depended on the trait in
question. In the case of larval growth rate, ramets and short shoots
accounted for most of the explained variation. In the case of biochemi
cal compounds, trees accounted for most of the variance in the content
of total phenolics and individual low-molecular-weight phenolics. In
the content of carbohydrates (total carbohydrates, starch, fructose, g
lucose, and sucrose) and amino acids, variation among branches was gen
erally larger than variation among trees. Variation among ramets was l
ow for most compounds. No single leaf trait played a paramount role in
larval growth. Secondary compounds, represented by phenolic compounds
, or primary metabolites, particularly sugars, may both be important i
n determining the suitability of birch leaves for larvae. If phenols a
re causally more important, genet-specific analyses of foliage chemist
ry are needed. If sugars are of primary importance, within-genet sampl
ing and analysis of foliage chemistry are necessary.