POPULATION VARIABILITY IN DANTHONIA CAESPITOSA (GRAMINEAE) IN RESPONSES TO INCREASING DENSITY UNDER THREE TEMPERATURE REGIMES

Abstract

The objective of this research was to compare the responses to increasing density of different natural populations of Danthonia caespitosa, a perennial grass of southern Australia, under different temperature regimes. Seeds were collected from four populations along a 167‐km N–S transect beginning north of Deniliquin, N.S.W., and ending near Heathcote, Victoria. Seedlings from each population were planted in loamy sand at three densities and grown under day/night temperature regimes of 15/10, 24/19, and 33/28 C in the Canberra Ceres Phytotron. Results indicate that one natural population (or one temperature regime) can not be used to characterize this species' responses to increasing density. The effects of density and temperature on tillering, plant height, leaf length, and leaf width varied significantly among populations, so that populations achieved comparable shoot weights by different relative responses to density. Variability in biomass among individuals (as indicated by coefficients of variation) increased from low to intermediate density; however, size variability at the highest density was either greater, lesser, or approximately the same as that of plants at the intermediate density, depending upon the population and the temperature regime. The combinations of density and temperature which produced the maximum shoot growth per pot varied markedly among populations; at 24/19 C, populations from frequently disturbed areas of low perennial plant cover had greater shoot weights per pot at the intermediate density, while those populations from more stable grassland sites had their greatest shoot weights at the highest density. Flowering occurred in all populations at the low density in 15/10 C; however, only for the population from the ungrazed grassland was there any flowering in the other temperature regimes or at higher densities. It is concluded that a “response to density stress” could be differentially described for each population at each temperature regime, even though the four study populations represented only 1/10 of the latitudinal range of D. caespitosa.