Experimental design
We conducted a microcosm experiment with three-factor factorial design. The first factor was target species, including S. grandis andS. krylovii . The second factor was soil nutrition condition with low nutrition treatment (no nutrition addition) and high nutrition treatment (adding slow-release inorganic fertilizer twice). The third factor was plantation condition, with S. grandis and S. kryloviigrown alone and in two-species mixtures (each in mixture with L. chinensis and A. cristatum , respectively, and in S. grandisS. krylovii mixture system). The target species grown alone was used for calculating the intensity and importance of competition. Therefore, the factor of neighbor species was viewed as the third factor when we analyzed the effect on the intensity and importance of competition in order to avoid ambiguity with “plantation condition”.
The seeds of each species, S. grandis , S. krylovii ,L. chinensis and A. cristatum, germinated in sterile sand, then seedlings of similar sizes were transplanted into the pots on July 4, 2011. Each pot with 20-cm depth and 10-cm diameter was filled with 1.5-kg soil. The high nutrition treatment was added 5g/pot slow-release inorganic fertilizer (15% N, 9% P2O5, 9% K2O) at the beginning and the middle of the experiment. For the treatment of target species grown alone, only one seedling were planted in each microcosm, and for the treatment of two-species mixture, one seedling per species was planted with the distance of 2 cm. Each treatment was replicated six times, totaling 168 microcosms used in this experiment. The location of each microcosm was randomly changed every week to avoid position effect. The experiment was carried out in the experimental field at Nankai University, and lasted 19 weeks. During the experiment, weeds in the microcosms were removed regularly, and shading, water stress, and light stress were avoided. At the end of the experiment, the individuals in each microcosm were obtained by species, dried to a constant mass and weighed. The root: shoot ratio which is a common response to the competition for soil nutrient and light was calculated (Reynolds et al. 1993; Robinson et al, 2010).