Time course of plant-plant interactions in experimental mixtures of annuals: density, frequency, and nutrient effects

Stellaria media and Poa annua were grown in a greenhouse over a wide range of mixed densities, and at three nutrient levels. The relative performances and interactions between individuals of both species were assessed over four interharvest growth periods using a new approach based on response functions (Connolly 1987). Species relative performances and interactions were complex, varying through time, and with mixed density and the nutrient environment. Studying the dynamics of the individual, interharvest growth periods led to a different, more accurate mechanistic interpretation of the final yield results, than did studying only one overall period of growth. Starting from considerably smaller seeds, Stellaria averaged twice as much biomass as young Poa plants by harvest 2 (H2). Stellaria achieved this early dominance with little suppression of Poa. Interspecific interference became more significant, and species proportional growth (K) became more similar in the second interharvest period (H2-3). During the final period of growth (H3-4), species relative performances largely reversed, Poa showing greater proportional growth than Stellaria. This superior performance by Poa occured despite individual plants being, on average, one half the size of Stellaria. Lack of interference early on is attributed to complementary above ground resource use resulting from species different shoot architectures (prostrate vs. cespitose). This complementarity disappeared as plants aged and became larger. The decline in Stellaria's superior relative growth performance through time was mainly due to its earlier and more substantial diversion of resources to reproduction. Species perception of each others influence, relative to themselves, as quantified by substitution rates, varied with both mixed density and nutrient environments. Stellaria consistently perceived Poa as less influential at higher nutrient levels than at lower levels. High nutrients favored the production of biomass for Stellaria more than for Poa, particularly in H2-3. While accelerating the switch to reproduction for Stellaria, nutrients did not increase its final reproductive yield per unit biomass. After H2, proportional growth for both species was relatively independent of frequency and density, indicating that a species acted as an aggregate of similarly behaving units of biomass whether arranged on many small, or a few large plants. An example shows that the conclusions from the response function approach applied within can be qualitatively different from those derived from a substitutive approach such as replacement series.