CROWN CONSTRUCTION, LEAF DYNAMICS, AND CARBON GAIN IN 2 PERENNIALS WITH CONTRASTING ARCHITECTURE

We used two co-occurring plant species of similar size and leaf morpho logy but with contrasting architecture to test if they would show spec ific patterns of leaf demography and leaf ecophysiology when they were grown in a range of experimental environments. Aster lanceolatus, a s pecies with branched shoots, had exponentially increasing leaf populat ions. This led to high leaf turnover rates and to crowding of leaves a t the top of the plants. Solidago canadensis, a species with unbranche d shoots and rapid height growth, had linearly increasing populations of leaves, leading to a uniform vertical distribution of leaves and to lower leaf turnover rates. In comparison with the short-lived leaves of A. lanceolatus, the leaves of S. canadensis lived longer, contained more nitrogen and calcium per dry mass, and responded to changing lig ht conditions within the plant crown by adjusting their angles towards the sun. Leaves of A. lanceolatus had an early peak in gas exchange a ctivity followed by a rapid decline, whereas gas exchange activity in leaves of S. canadensis at the beginning of leaf development was lower than in A. lanceolatus but decreased only slightly with leaf age. Wit hin the narrow vertical band in which leaves of A. lanceolatus were co ncentrated, young leaves at the top had higher mass per area, higher c hlorophyll alb ratios, and much higher rates of photosynthesis and con ductance than the older leaves beneath. In S. canadensis, in which lea ves were distributed over much of the vertical dimension along the pla nt, upper leaves, in contrast to lower (older) leaves, had lower mass per area, were held less perpendicular to the sun, and had higher rate s of photosynthesis in high but not in low light. Because leaf populat ions grew exponentially in A. lanceolatus, assimilates could be re-inv ested; thus leaf number during the second half of the growing season w as a good predictor of final biomass. In S. canadensis, with linearly growing leaf populations, the assimilates that were not re-invested in new leaves were allocated to the stem, and stem height or volume were good predictors of final biomass. Compared with the strong influence of architecture on the structure and dynamics of leaf populations, the effects of background species and fertilizer application were weak bu t nevertheless significant. Background species of similar aboveground mass but different stature (the tall Solidago altissima and the short Poa pratensis) affected leaf deployment (internode elongation, leaf or ientation, and leaf survival), probably via competition for light, but not leaf quality (morphology, physiology) of the two target species A . lanceolatus and S. canadensis. Further, diameter-height allometries, which varied considerably among plants, tended to be flatter if the t arget species were grown in the tall than in the short background. Fer tilizer application accelerated plant growth and phenological developm ent. It increased light competition among leaves and leaf turnover; ne vertheless, the photosynthetic water use efficiency of old leaves was higher in fertilized than in unfertilized plants. We suggest that the characteristic leaf dynamics of A. lanceolatus and S. canadensis are t ypical for the contrasting aboveground architectures these species rep resent. Based on the results of this and previous studies we suggest t hat an important driving force in the evolution of these complex adapt ations is the degree of mixing within canopies of leaves of different plant genotypes. Leaves (and branches) are expected to be more autonom ous if they frequently interact with leaves from other genotypes, as i n the polyclonal patches of A. lanceolatus with its long and interming ling belowground rhizomes, than if they usually interact with leaves f rom the same genotype, as in the monoclonal patches of S. canadensis w ith its short rhizomes and compact belowground architecture.