Abstract The effects of competition, grazing, and microsites on seedling dynamics of Festuca spp. were investigated in a semiarid steppe of Patagonia, Argentina. In an exclosure and an adjacent grazed area, the level of root competition for seedlings was controlled through root exclusion tubes (0.1 m diameter, 0.4 m long) installed adjacent to seven similarly-sized Festuca pallescens plants and in the interspace between F. pallescens plants. Seedling dynamics at the phenological stages of a) recently emerged and up to four leaves and b) five leaves and up to one tiller were followed inside the tube area (no competition) and in paired circles (competition) for approximately three growing seasons. Peak density of Festuca spp. seedlings occurred in fall and early spring, when water content in the first 0 to 5 cm of the soil was above 8%. Seedling densities of Festuca spp. at the two phenological stages were significantly reduced by grazing. At the phenological stage of one to four leaves, seedlings were not affected by root competition, although their densities were significantly higher for seedlings adjacent to F. pallescens plants than in the interspace between plants. This was due to higher soil water content in these areas throughout the season in the 0 to 5 cm of the soil. Seedling densities at the phenological stage of five leaves and up to one tiller were significantly higher when there was no competition from neighboring grass plants as compared to seedlings exposed to competition. In regard to microsite differences, seedling density was significantly greater in the interspaces than adjacent to mature plants. These results suggest that although higher seedling densities can be obtained by excluding the area from grazing, intraspecific competition for soil water during the summer drought period can act as a barrier for further Festuca spp. seedling establishment.

This study was funded by a grant from the Consejo Nacional de Investigaciones Científícas y Técnicas (CONICET) de la República Argentina (PID 3058100/88). Additional support was provided by the Idaho Forest, Wildlife and Range Experiment Station (Contribution no. 793).