All gas exchange measurements were made on mature Juniperus ashei leaves on the south side of trees on the western part of the University of Texas at San Antonio campus in central Texas (29.58030 N and −98.62403 W). Area topography is rolling with low slopes between 4.5˚ and 13.5˚ and soils that are clayey-skeletal, smectitic, thermic lithic calciustolls [38] in the Tarrant association with surface horizons between 0 and 25 cm thick [39] . The area geology is composed of heavily fractured limestone over limestone bedrock. Climate is subtropical sub-humid [40] having a mean annual temperature of 20˚C with means of 9.6˚C in January and 29.4˚C in July [41] . Precipitation is also highly variable usually 78.7 cm/yr and bimodal, very little in June and July and peaks in May 10 with 7 cm and September with 8.7 cm [41] . There was no domestic grazing or domestic livestock present in the study area for the past 75 years. There are large areas of Juniperus ashei/Quercus virginiana woodlands or savannas on former grassland sites in central Texas which are considered representative of similar communities found in this area [42] .

Mature J. ashei plants in relatively undisturbed J. ashei/Q. virginiana woodland communities were selected randomly for measurements. Mean density and basal area of the communities examined was determined but limited information about community structure is presented here [2] [33] . Trees found were all identified, counted and measured. Concise but succinct information about community structure is presented below for the major species found.

Gas exchange responses at both ambient and elevated levels of CO₂ and temperature were made. Three plants were randomly selected in the summer of 2007. On each replicate plant one main stem tip at breast height, approximately 137 cm above the soil surface was selected and only mature, non-damaged stems were used. Steady state photosynthetic light response curves (A_net vs. PPFD) were measured on full grown stem tips with completely expanded leaves at mid-day (1000 - 1400 hrs) when relative humidity had stabilized [43] . Leaves one cm from the growing stem tip were 1.07 ± 0.27 mm (mean ± 1.0 SD) in length and those five cm from the growing tip were 2.17 ± 0.34 mm in length and leaves surrounded the entire stem tip.

To measure leaf gas exchange, a branch tip or cluster with several secondary branches all with many small leaves was placed into the cuvette chamber. The stems were all parallel in the chamber in a single plane configuration and all at 90˚ to the light source. At least 66% of the cuvette chamber was covered. After the gas exchange measurements were made Leaf area within the chamber was measured with a LI-COR LI 3000A portable area meter. Each sample was measured in the scanner three times and the mean was determined and used as a correction and manually entered into the Li-6400® to adjust each curve. Measurements made and recorded were: A_net (net photosynthesis = µmolCO₂·m⁻²·s⁻¹), C_i (intercellular [CO₂] = µmolCO₂·molair⁻¹), T_leaf (chamber leaf temperature = ˚C), T_air (air temperature outside the chamber = ˚C), PPFD (photosynthetic active radiation = mol·m⁻²·s⁻¹), g (stomatal conductance = molH₂O·m⁻²·s⁻¹) and E (transpiration = mmolH₂O·m⁻²·s⁻¹).

The gas exchange chamber was used to mimic varying degrees of environmental modifications with the stem cluster in the chamber attached to the plant. Conditions adjusted or manipulated were the light level, CO₂ concentration, and temperature. Relative humidity was kept at 30% - 40% and the gas flow rate was set at 400 µmol·s⁻¹. Coefficient of variation stabilized (<1%) before recording and moving to the next measurement. Light levels started at 1800 µmol·m⁻²·s⁻¹ and decreased to 1600, 1400, 1200, 1000, 800, 600, 400, 200, 100, 75, 50, 25, 10, 5 and finally 0 µmol·m⁻²·s⁻¹. Light curves and CO₂ response curves were measured for different combinations of the leaf chamber CO₂ and temperature environments.

The leaf chamber was first set at the mean atmospheric CO₂ level (390 µL/L) for 2007 and a temperature of 35˚C. This temperature was chosen based on the mean high temperatures for San Antonio during the summer months of June, July and August. Light curves were repeated holding the ambient CO₂ constant while raising the chamber temperature to 40˚C and then to 45˚C. Next, the leaf chamber CO₂ was raised to 1.5 times the 2007 CO₂ levels to 585 µL/L. Light curves were completed at a temperature of 35˚C, 40˚C and 45˚C. This process was then repeated with the leaf chamber CO₂ level set at twice the 2007 ambient level at 780 µL/L. Lastly, CO₂ response curves were measured at a canopy shade light level (700 µmol·m⁻²·s⁻¹). Measurements were made at 35˚C, 40˚C and 45˚C.

The data analysis was completed using Microsoft Excel^© and JMP^© IN 5.1. Significant differences were measured using the JMP^© IN 5.1 software with a repeated measures MANOVA on the photosynthetic rate curves including intercellular CO₂ concentrations, stomatal conductance and transpiration. The light level, PPFD, was the repeat variable [44] . Water use efficiency (WUE) was calculated by dividing the photosynthetic rate by the transpiration rate and also analyzed using a repeated measures MANOVA. Significance levels used for all tests were P ≤ 0.05. Normality was checked with the Shapiro-Wilk W test and homogeneity of variance with Bartlett’s test and log transformed as necessary. A standard least squared ANOVA was used to detect significant differences in each curve at each CO₂ concentration and temperature combination. However, this is a curve-to-curve comparison and individual CO₂ uptake was not compared at individual light levels on each plant and each replication.

For other measurements, they were derived from Excel^® plots of the LICOR^® Li-6400 measurements. Included were maximum photosynthetic rate (A_max) which was the highest A_net measured for each replicate or a mean of the highest A_net values that were not significantly different. The dark respiration rate (R_d) was the gas exchange rate at PPFD = 0 µmol·m⁻²·s⁻¹. The quantum yield (Ø) was the linear initial slope relationship calculated using the dark values and A_net at increasing PPFD until the regression coefficient of the slope decreased. The light compensation point (L_cp) was calculated as the PPFD when A_net = 0 µmolCO₂·m⁻²/s⁻¹ using the linear regression of the initial response. The light saturation point (L_sp) was the light level when the initial slope reached A_max. A standard least squared ANOVA was used to determine significant differences for the CO₂ concentration and temperature effects. Tukey-Kramer HSD multiple comparison tests were used to determine differences between pair wise comparisons [44] .

The community overstory had a mean canopy density of 1840 plants/ha and was found to be dominated by Juniperus ashei with a relative density in the canopy of 61% ± 12% (mean ± one standard deviation). The other major canopy species was Quercus virginiana with a relative canopy density of 36% ± 6%. Additional community species with relative densities of 0.06% - 1.80% were Celtis laevigata (sugarberry or hackberry), Diospyros texana (Texas persimmon), Prosopis glandulosa (mesquite), Calia secundiflora (Texas mountain laurel), Ulmus crassifolia (cedar elm), and Ungnadia speciosa (Mexican buckeye).

The mean curves of the photosynthetic rates for Juniperus ashei are shown by temperature and CO₂ concentration with light level being fixed ( Figure 1(A) and Figure 1(B) ). The photosynthetic rates by temperature show a statistical difference (MANOVA, P = 0.0452). When the temperature was increased there was a non-significant decrease in photosynthetic rate by approximately 9% between the 35˚C and 40˚C curves and an additional non-significant decrease of 13% between the 40˚C to 45˚C (P = 0.4104 and P = 0.2876 respectively). However, there was a significant decrease in photosynthesis between the 35˚C and the 45˚C curves (P = 0.0446, Figure 1(A) ).

The comparison of the mean photosynthesis rates by CO₂ levels shows a significant difference between the three curves at different CO₂ levels (MANOVA, P = 0.0003, Figure 1(B) ). The total increase was 51% for the curves as the CO₂ concentration increased. The CO₂ curves showed a similar significant difference between the ambient (390 µL/L) or low CO₂ concentration and the middle (585 µL/L) concentration (≈25% increase, P = 0.0306) and the middle and high (780 µL/L) CO₂ concentration (≈26% increase, P = 0.0318).

Repeated measures MANOVAs were used to investigate the main effects of light levels, CO₂ concentration and temperature on Juniperus ashei. Because the interactions were not significant except for water use efficiency they were removed from the MANOVA ( Table 1 ). For J. ashei as light levels were increased all factors measured changed as the temperature and the CO₂ concentrations were increased. However, there were no significant temperature effects on; conduction, intercellular [CO₂] concentrations, or transpiration ( Table 1 ). Temperature only had a significant effect on the photosynthetic rates and water use efficiency (WUE). For CO₂ concentration there were significant effects on photosynthetic rates, stomatal conduction, intercellular [CO₂] concentration, and WUE. Transpiration rate was not significantly affected by increased temperature or CO₂ ( Table 1 ).

The maximum photosynthetic rate (A_max) did not change with temperature (ANOVA, P = 0.1468, Table 2 ). Temperature had little effect on the mean A_max although it did drop by 18% with a standard error of 0.92 µmolCO₂·m⁻²·s⁻¹. The A_max did change significantly with CO₂ concentration (ANOVA, P < 0.0001, Table 2 ). In addition, Tukey comparisons of the CO₂ effect showed significant differences between each concentration. Elevating CO₂ increased A_max from 9.42 µmolCO₂·m⁻²·s⁻¹. for the ambient CO₂ concentration to 12.97 µmolCO₂·m⁻²·s⁻¹. for the middle CO₂ concentration and to 17.20 µmolCO₂·m⁻²·s⁻¹. for the high CO₂ concentration. This was a total A_max increase of approximately 45% from the low to high CO₂ concentration. The interaction term was not significant (P = 0.5961, not shown).

The light saturation point (L_sp) was not significantly different by temperature (ANOVA, P = 0.8776, Table 2 ). The mean L_sp for the three temperatures were different by 5% with a standard deviation of 23.8 µmolCO₂·m⁻²·s⁻¹. Elevating CO₂ had a significant effect on L_sp (ANOVA, P = 0.0001, Table 2 ). A Tukey comparison showed a significant difference between values for the ambient CO₂ L_sp and both the medium CO₂ and high CO₂ L_sp. There was no significant difference for L_sp between the middle and the high CO₂ concentrations. The interaction term was not significant (P = 0.8929, not shown).

The light compensation point (L_cp) showed a significant difference by temperature (ANOVA, P = 0.0003, Table 2 ) but not by CO₂ concentration (ANOVA, P = 0.3729, Table 2 ). A Tukey comparison showed the 35˚C L_cp was not significantly different than the 40˚C L_cp. The 45˚C L_cp was significantly different from both the 35˚C L_cp and the 40˚C L_cp. There were no significant differences in the L_cp for CO₂ concentration (ANOVA, P = 0.3729). The interaction term was not significant (P = 0.5166, not shown).

The ANOVA for the dark respiration rates (R_d) showed a significant difference by temperature but not by CO₂ concentration (P = 0.0007 and P = 0.4446, Table 2 ). Tukey comparison showed the 35˚C R_d was significantly different from the 45˚C R_d while the 40˚C R_d was not significantly different from either the 35˚C or the 45˚C R_d. The CO₂ concentration showed no trend with R_d values with increasing CO₂ concentration (ANOVA, P = 0.4446). The interaction term was not significant (P = 0.4226, not shown).

The quantum yield (Ø) showed no statistical trend based on temperature (ANOVA, P = 0.6881, Table 2 ). As with temperature, the CO₂ ANOVA showed no significant effect on the Ø values (P = 0.2075, Table 2 ). The interaction term was not significant (P = 0.9163, not shown).

The mean curves of the water use efficiency (WUE) for J. ashei are shown by temperature, CO₂ and light effects ( Figure 2(A) and Figure 2(B) ). Water use efficiency was significantly different when compared by temperature (MANOVA, P < 0.0001, Figure 2(A) ). Water use efficiency values decreased from a plateau of approximately 3.5 mmol·mol⁻¹ to approximately 2.5 mmol·mol⁻¹ as temperature increased to 45˚C or by a total of 21%. The comparisons by CO₂ concentration were also statistically significant between the curves (repeated measures MANOVA, P < 0.0001, Figure 2(B) ). The curves generally increased as the light levels increased and as the CO₂ concentration increased. At the ambient (390 µL/L) or low CO₂ concentration the WUE value was lowest at approximately 2.1 mmol·mol⁻¹ ( Figure 2(B) ). At the highest CO₂ concentration, the WUE value increased approximately 50% to a value of approximately 4.2 mmol·mol⁻¹.

Because there was a significant interaction between light level, temperature, and CO₂ concentration for the WUE response an additional figure is presented to demonstrate the effect ( Figure 3 ). The WUE increased with light level (PFD, right axis front to back) and with CO₂ concentration (light, dotted to gray shade within each temperature level), but decreased overall with each temperature (X-axis-left to right).

Figures that represent the photosynthetic rate and water use efficiency CO₂ response to temperature are shown below ( Figure 4(A) and Figure 4(B) ). Both the CO₂ photosynthetic response and water use efficiency increase with increasing CO₂ concentration. Neither curve seems to reach a plateau. The curves follow the same basic trend, with no significant difference to increasing temperature. None of the repeated measures MANOVAs performed on the photosynthetic response, intercellular [CO₂], stomatal conductance, and transpiration were significantly different by temperature (Data not shown; P = 0.3405, P = 0.4603, P = 0.1345, P = 0.9629, respectively).