The characteristic decline in height growth that occurs over a tree's lifespan is often called "age-related decline." But is the reduction in height growth in aging trees a function of age or of size? We grafted shoot tips across different ages and sizes of Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) trees to determine whether the decline in height growth is mediated by tree size or by the age of the apical meristem. We also evaluated whether reduced carbon assimilation plays an important role in height growth decline. In one experiment we cut shoot tips from old-growth, young-mature and seedling trees and grafted them onto 2-year-old graft-compatible rootstock in a seed orchard in Lebanon, Oregon. In another experiment we performed reciprocal grafts between lateral branches of old-growth trees accessible from the canopy crane at Wind River, Washington and young-mature trees in a nearby plantation. We measured growth (diameter and elongation of the dominant new stem) and mortality annually for three years in the Seed Orchard experiment and for two years in the Reciprocal Graft experiment. In the Seed Orchard experiment we also measured photosynthetic capacity (determined from the response of net carbon assimilation to the intercellular CO(2) concentration of the leaf, or A/C(i) curves), leaf mass per area (LMA) and carbon isotope composition (delta(13)C) of cellulose in 1-year-old foliage. Grafting caused changes in both growth and physiology of the grafted stems. Within two years after grafting, growth and physiology of all combinations of scions and rootstock exhibited characteristics of the rootstock. In some cases, the change in growth was dramatic-cuttings from old-growth trees showed a 10-fold increase in stem elongation rate within 2 years of grafting onto seedling rootstock. Similarly, carbon isotope discrimination of new foliage on shoots from old-growth trees increased by nearly 3 per thousand and 2 per thousand after grafting onto young-mature and seedling rootstock, respectively, whereas discrimination decreased by a similar magnitude in scions from young-mature trees after grafting on old- growth trees. Furthermore, differences in carbon assimilation estimated from carbon isotope discrimination and A/C(i )relationships were small relative to growth differences. Our results confirm that size, not age, drives developmental changes in height growth in Douglas-fir. Reduced carbon assimilation does not play an important role in height growth decline.