The rate at which tree height increases slows as trees grows taller.  For example, 60-m-tall Eucalyptus regnans and 70-m-tall Sequoia sempervirens both grow 43 cm yr^-1 in height in 2006, but 90- and 110-m-tall trees of the same species at the same sites grew only 27 and 8 cm yr^-1, respectively.  Tree growth, however, is far more than an increase in height, because growth occurs across the entire surface of a tree’s cambium.  Accurate quantification of whole-tree dimensions has now been achieved for many individuals of the 3 tallest tree species (S. sempervirens, E. regnans, and Pseudotsuga menziesii).  For example, all branches, limbs, and trunks were non-destructively measured for size, structure, and location on 70 P. menziesii trees 50 to 650 years old in Washington.  Principal components analysis of 24 tree-level variables revealed two orthogonal dimensions of structure that accounted for 71.3 and 12.4 % of the total variation.  The first dimension represented a gradient of overall tree size and structural complexity that was positively correlated with stand age (R² = 0.99).  The second dimension represented a gradient of tree growth potential that was positively correlated with measured growth of main trunks at breast height during the 5-yr period after trees were mapped (R² = 0.39).  The strongest determinant of growth potential was the ratio of a tree’s total cambium surface area to projected leaf area.  A tree’s growth potential may thus be unrelated to either size or structural complexity and depend instead on factors related to whole-tree carbon balance.