The survival of grasslands is often limited by summer drought. The objective of this study was to assess the effects of drought stress on leaf physiological traits of switchgrass (Panicum virgatum L.) "Cave-in-Rock" and "Alamo", big bluestem (Andropogon gerardii Vitman) "Pawnee", indiangrass (Sorghastrum nutans (L.) Nash) "Nebraska 54" and bermudagrass (Cynodon dactylon L.) "Cheyenne". Three water regimes (100%, 50% and 30% of field capacity) were maintained by watering every 2-d to constant soil water content in a greenhouse from April through December in 2003 and 2004. There were 3 replicates of each treatment combination of grass species and water regime (72 pots in total).

Midday water potential (WP) of control treatments differed (P<0.05) from water stressed treatments 61 d (2003) and 46 d (2004) after withholding water. Bermudagrass had lower (P<0.05) water potential than other species. Declines in WP were correlated with osmotic potential (OP) (r = 0.96; P<0.05). Big bluestem and switchgrass had higher (P<0.05) cell membrane stability (CMS) values than indiangrass and bermudagrass in both years. The CMS was uncorrelated with relative water content (RWC)in both study years (r = 0.072 in 2003; r = 0.084 in 2004). Significant differences (P<0.05) in percent ion leakage ratios (L) were found 100 days (2003) and 62 days (2004) after drought initiation. Leaf proline concentration of stressed plants increased 23-fold (2003) and 25-fold (2004) over controls. Switchgrass and big bluestem had significantly greater (+21%, P<0.05) proline responses than indiangrass and bermudagrass at the end of the experiment.

We concluded that leaf responses to drought are mediated through changes in osmotic potential and that reduced soil water had less direct effects on early leaf physiological responses and was more closely correlated with long term leaf physiological responses such as proline accumulation and ion leakage.