EFFECTS OF NUTRIENT AVAILABILITY AND COMPETITIONBETWEEN TWO GRASS SPECIES LEYMUS CHINENSIS (TRIN.)TZVEL. AND PHRAGMITES COMMUNIS TRIN. ON THE PLANTPERFORMANCE

摘要

Leymus chinensis (Trin.)Tzvel., a perennial grass, is a dominant speciesfrom arid to semi-arid steppes in northern Chi-na and eastern Mongolia. Phragmites communisTrin. is also a perennial grass, and is distributedwidely in the world. In the natural grasslands ofnortheastern China, both species always co-existas co-dominating species due to their commoncharacteristics such as propagation both by seedsand vegetative reproduction. Replacement seriesexperiments were used to test the effects of nutri-ent availability and competitive interaction on thegrowth performance of two clonal plant species.The experimental treatments included five nutri-ent levels (3.6, 7.2, 10.8, 14.4 and 18.0 kg organicmatter per pot, 20 cm diameter and 15 cm deep)and five species proportions (20:0, 16:4, 10:10,4:16 and 0:20 for L. chinensis and P. communis,respectively) with twenty tillers in total per pot.Each treatment had 10 replications. Growth char-acteristics including tiller height, tiller number,plant biomass, rhizome length and bud numberof plants in monoculture and mixture culturewere recorded and compared to examine the ef-fects of nutrient and competitive interaction onthe plant performance. The growth of L. chinensisand P. communis in mixture was influenced bythe nutrient availability and competition, whichdepended on the combination between nutrientlevel and species proportion. The results impliedthat the intensity of competition should be lowerin nutrient-poor habitats when the co-existing species demanded on the same limiting resource.R communis benefited from coexisting with L.chinensis, especially under nutrient-rich condi-tions. The aboveground relative yield (RYabove)expressed in units of tiller height, dry biomassand daughter tiller number was recommended asan effective and simple index to predict the rela-tive competitive ability for clonal plants. It wasbased on the regression for RY_(aove)and RY_(below)(thebelowground relative yield) against RY (relativeyield), measured as yield in mixture divided bythat in monoculture.