Future changes in climate are likely to strongly affect plant physiological and growth parameters, and thus potentially influence competitive interactions among plants. We used rainout shelters and sprinklers to test the influence of changes in precipitation on aboveground physiological parameters and belowground growth of honey mesquite (Prosopis glandulosa) and black grama (Bouteloua eriopoda) at the Jornada Basin LTER. In this area of the northern Chihuahuan Desert, mesquite and other woody plants have encroached into former desert grasslands over the past 150 years. Changes in precipitation may be important in influencing future dynamics of this shifting vegetation. Five precipitation manipulations were initiated in 2007. Rainout shelters reduced rainfall by 50% or 80% relative to ambient precipitation, sprinklers increased rainfall by 50% or 80% relative to ambient precipitation, and control plots had no precipitation manipulation. We measured gas exchange (Amax and carboxylation rate), chlorophyll fluorescence, pre-dawn water potential (ψpd), and foliar N content during the first two growing seasons of the experiment (2007-8) and collected minirhizotron root images in August of the second growing season of the experiment (2008). Physiological responses to the treatments varied through time depending on weather patterns. There was a positive response of B. eriopoda ψpd to precipitation on sampling dates when plants were the most water stressed. However, B. eriopoda leaf N content and Amax responded positively to precipitation only at the end of the monsoon seasons when it was relatively wet. In contrast, P. glandulosa ψpd exhibited a positive response to precipitation treatments on sampling dates with least water stress and neither Amax nor leaf N responded to precipitation treatments. For both species, carboxylation rate responded positively to precipitation but there was no significant effect of precipitation on midday photoinhibition or evening recovery in photosystem II. Root area was highly variable among plots, and precipitation explained very little of this variation (r2 = 0.06). Rainfall for both years of the experiment was well above the long-term average (145% and 124% for 2007 and 2008, respectively). The very high ambient precipitation may have minimized the chance of seeing strong precipitation responses, as even the 80% precipitation reduction had relatively high soil moisture through much of the growing season. Longer term studies over additional years of precipitation manipulation will be necessary to explore how short and long term changes in ambient precipitation affect physiological and competitive processes.