Figure 1. Early-summer yields for well-watered (wet) and drought-stressed (dry) two-species grass–legume and three-species grass–legume–chicory mixtures. Means with the same letter are not significantly different at P = 0.05.
Figure 2. Late-summer yields for well-watered (wet) and drought-stressed (dry) two-species grass–legume and three-species grass–legume–chicory mixtures. Means with the same letter are not significantly different at P = 0.05.
Figure 3. Root distribution with soil depth for two- and three-species mixtures. (A) Initial distribution at the beginning of first drought-stress treatment (6 June 2003). (B) Distribution beneath well-watered plots on 1 July 2003. (C) Distribution beneath drought-stressed plots on 1 July 2003. Each point represents the mean of four replications and five frames (6.75 cm depth). Error bars represent ± 1 SE.
Figure 4. Effect of drought treatments on soil volumetric water content at 25-, 50-, and 75-cm depths in 2003. No significant differences existed between the two- and three-species mixtures in 2003 for either moisture treatment on any date.
Figure 5. Effect of drought treatments on soil volumetric water content at 25-, 50-, and 75-cm depths in 2004. * indicates dates when significant differences occurred between the two-and three-species mixtures under drought stress.
Figure 6. Soil water potential at 25-cm depth during the final 18 d of drought stress in 2003 for two-species (grass–legume) and three-species (grass–legume–chicory) mixtures. Standard errors are indicated by gray shaded area. Only negative standard error values are shown to improve clarity.