• Using the Poplar Free Air CO2 Enrichement (PopFACE) facility we investigated the effects of elevated [CO2] on the diurnal and growth cycle responses of photosynthesis and conductance in three poplar species.
• In situ diurnal measurements of photosynthesis were made on Populus alba, P. nigra and P. ×euramericana and, in parallel, in vivo maximum capacity for carboxylation (Vc,max) and maximum rates of electron transport (Jmax) were determined by gas exchange measurement.
• Light saturated (Asat) and daily integrated (A′) photosynthesis increased at elevated [CO2] in all species. Elevated [CO2] decreased Vc,max and Jmax for P. nigra and Jmax for P.¥euramericana but had no effect on stomatal conductance in any of the species throughout the first growth cycle. During post-coppice re-growth, elevated [CO2] did not increase Asat in P. nigra and P.×euramericana due to large decreases in Vc,max and Jmax.
• A 50% increase in [CO2] under these open-air field conditions resulted in a large and sustained increase in Asat. Although there were some differences between the species, these had little effect on photosynthetic rates at the growth [CO2]. Nevertheless the results show that even fast growing trees grown without rooting volume restriction in the open may still show some down-regulation of photosynthetic potential at elevated [CO2].
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
net daily integrated carbon assimilation (mol m−2)
light saturated rates of net carbon assimilation (µmol m−2 s−1)
leaf intercellular CO2 concentration (µmol mol−1)
atmospheric CO2 concentration
free air CO2 enrichment
stomatal conductance (mmol m−2 s−1)
maximum in vivo rates of electron transport through photosystem II (µmol m−2 s−1)
poplar free air CO2 enrichment
photosynthetically active photon flux (µmol m−2 s−1)
leaf temperature (°C)
maximum in vivo velocity of carboxylation (µmol m−2 s−1)
vapor pressure deficit (kPa)
maximum quantum efficiency of carbon assimilation (mol CO2 mol−1Q)
The Intergovernmental Panel on Climate Change (Prather et al., 2001) predicts that atmospheric CO2 concentration ([CO2]) will increase 50% by the middle of this century. The scale of this response is unprecedented in recent geological history; therefore, it has become increasingly more important to address the effect it will have on vegetation. Because photosynthesis is the primary process by which carbon enters the biosphere and by which plants sense rising [CO2], it is critical to understand how it will respond to an elevated [CO2] environment (Drake et al., 1997). Forest tree species are a critical component of the global carbon budget accounting for over half of the total net carbon uptake into terrestrial vegetation (Geider et al., 2001); therefore, it is important to address how photosynthesis for tree species will respond in growth in elevated [CO2].
Despite the wealth of previous studies, major limitations exist on information for trees grown in elevated [CO2]. First, most studies have used enclosures which often show a larger effect on the vegetation than the treatment itself (Allen et al., 1992; McLeod & Long, 1999). Recent literature reviews also suggest that photosynthetic responses to elevated [CO2] may be highly dependent on fumigation method (Curtis, 1996; Curtis & Wang, 1998; Ainsworth et al., 2002). Secondly, most studies concern juvenile individuals and are conducted before canopy closure; yet most terrestrial carbon assimilation is by mature trees in closed canopies (Lee & Jarvis, 1995; Norby et al., 1999). The use of Free Air CO2 Enrichment (FACE) technology provides the opportunity to determine plant responses to elevated [CO2] under field conditions without any direct alterations in the canopy microclimate (Hendrey et al., 1993; Ellsworth et al., 1997; McLeod & Long, 1999; Gunderson et al., 2002) and allows trees to grow to canopy closure. Recent studies that address photosynthetic responses of tree species to growth in FACE systems generally show a lack of photosynthetic acclimation responses (Tognetti et al., 1999; Herrick & Thomas, 2001; Gunderson et al., 2002); although pine trees exposed to elevated [CO2] using FACE have shown some down-regulation (Ellsworth et al., 1997). Two such FACE facilities, the Duke Forest loblolly pine and the Oak Ridge sweetgum FACE experiments, began fumigation when the trees were already mature and the canopy was at or close to closure (Ellsworth et al., 1997; Gunderson et al., 2002). Stand development, growth and canopy closure at elevated CO2 might result in different responses than those observed from facilities where CO2 fumigation began after canopy closure.
The Poplar Free Air CO2 Enrichment (PopFACE) experiment (Miglietta et al., 2001) provides a unique opportunity to address how photosynthetic responses to elevated [CO2] are affected by growth from planting to canopy closure and to harvesting under open-air elevation of [CO2]. Poplars are so fast growing that they provide a rare opportunity to grow a plantation forest from planting to canopy closure of tall trees (> 9 m) in just 3 yr. Because PopFACE is a coppice system, it is also possible to determine the effect of elevated [CO2] on re-growth immediately after coppicing when the source-sink balances in these species have been significantly altered. There were two objectives of this study, to determine: diurnal and growth cycle responses of photosynthesis and conductance for three poplar species grown in elevated [CO2]; and the response of leaf photosynthesis to elevated [CO2] during postcoppice re-growth.
Materials and Methods
The PopFACE facility is located near Viterbo in Central Italy (Tuscania; 42°22′-N, 11°48′-E, alt. 150 m). The 9 ha field was originally used for wheat cultivation and consists of a heavy loam soil. The entire field was planted with a poplar hybrid, Populus×euramericana Dode (Guinier) (P. deltoides Bart. ex Marsh. × P. nigra L., I-214) at 2 m × 1 m spacing (5000 trees ha−1), with the exception of six 30 × 30 m square plots. Within each plot, a 22-m diameter ring which included about 350 individual trees was established with three such rings serving as controls (370 µmol mol−1) and three receiving [CO2] enrichment (550 µmol mol−1). Within each elevated plot, pipes were installed in a 22-m diameter ring in order to elevate [CO2]. The pipes release CO2 in a precisely controlled manner according to wind speed and direction to achieve a uniform elevation of [CO2] within the three treatment plots. The construction and performance of this Free-Air CO2 Enrichment (FACE) system is described in detail by Miglietta et al. (2001). Each ring was divided into six equal segments, pairs of opposing segments were planted to a single clone of each of three poplar species P. alba L. (genotype 2AS11), P. nigra L. (genotype Jean Pourtet) and P. × euramericana (genotype I-214). Planting density within the experimental plots was 1 m × 1 m spacing (104 trees ha−1). A drip irrigation system was installed to prevent drought-stress and was used regularly to maintain soil moisture. Detailed information on plant material and on plantation layout is presented in Scarascia-Mugnozza et al. (2000) and Calfapietra et al. (2001). CO2 fumigation began immediately after planting in 1999 and was maintained from bud burst (March) until leaf senescence (November) of each year. The trees were coppiced between the end of October 2001 and February 2002 and the fumigation was continued in March 2002 for the re-growth.
Gas exchange measurements
A vs Ci response Measurements of A vs Ci were first performed in late August, 1999 and were repeated during the months of May, July and September, for the 2000 and 2001 growing seasons and in June 2002, following the coppice. The youngest fully expanded mature leaves from two leader branches per species per plot were collected predawn and placed in individual 0.5 l containers in the dark. The petioles of the leaves remained under water from the time of cutting until the measurements were completed. It should be noted that cutting the leaves predawn and analyzing them in a controlled environment measured the maximum potential response of leaf photosynthesis. This procedure avoided photoinhibition, water stress or triose-phosphate utilization limitation that might develop over the diurnal course and ensured that differences reflected long-term acclimation rather than short-term diurnal effects. Stomatal conductance and rates of photosynthesis measured in these leaves equaled or exceeded rates measured in situ, suggesting that this procedure did not cause any loss in photosynthetic potential.
Gas exchange measurements were made using a portable gas exchange system (Li-Cor 6400; Li-Cor, Inc., Lincoln, NE, USA). The gas exchange system was zeroed daily using anhydrous calcium carbonate (Drierite, W.A. Hammond Drierite Company, Ltd, Xenia, OH, USA) to remove water and using soda lime (sofnolime granules, Morgan Medical, Ltd, Kent, UK) to remove CO2 from the air entering the cuvette. Leaf temperatures were set at 25°C for all measurements, though actual temperature ranged from 25 to 30°C. Leaves were illum-inated using a red-blue light source attached to the gas-exchange system and photosynthetically active photon flux density (Q) was maintained between 1000 and 1500 µmol m−2 s−1, depending on measurement date, for the duration of the A vs Ci response curve. Levels of Q varied based on saturating light levels derived from photosynthetic light response curves measured before experimental sampling. By using light levels slightly above saturating, we were able to reduce the chances of inducing photoinhibition during the measurements. Leaf vapor pressure deficits were maintained between 0.5 to approx. 1.6 kPa. Measurements of A were made starting at 400 µmol mol−1 CO2 surrounding the leaf, decreased stepwise to 50 µmol mol−1, returned to 400 µmol mol−1, and increased stepwise to 1600 µmol mol−1 CO2. Each complete curve consisted of at least eight separate measurements. Values for A and Ci were calculated using the equations of von Caemmerer & Farquhar (1981) and were used to solve for Vc,max and Jmax using the equations of Farquhar et al. (1980). When necessary, measurements were corrected to 25°C using the temperature responses of Bernacchi et al. (2001) and Bernacchi et al. (2003) for the Rubisco and RuBP-limited portions of the A vs Ci curves, respectively.
Asat Sampling and measurements of A vs Q curves were performed in September 2000 and in May, July and September during the 2001 growing season and in June 2002, following the coppice. Leaves were sampled as described for A vs Ci measurements above. Leaves were placed in the cuvette and illuminated until steady-state light-saturated rates of photosynthesis and stomatal conductance were achieved; typically this required 5 min. The [CO2] surrounding the leaf was set to the growth [CO2] for each treatment. Plots of A vs Q were then measured starting at saturating light (2000 µmol m−2 s−1) and decreasing stepwise to complete darkness. Curves consisted of a minimum of 13 measurements at different Q and were usually completed within three hours of leaf collection. Completed A vs Q curves were fitted to a nonrectangular empirical function to estimate light saturated photosynthesis (Asat). The apparent maximum quantum efficiency of CO2 assimilation () was calculated as the slope of A measured at a range of Q below 100 µmol m−2 s−1 where data points did not deviate from linearity (Long et al., 1993). For each measured curve, values of stomatal conductance (gs) were taken from the photosynthesis measurement made at a Q of 1500 µmol m−2 s−1, which was always saturating.
The diurnal response of leaf photosynthesis was measured three times during the 2000 growing season, mid-May, mid-July and mid-September. Beginning predawn and finishing after dusk, photosynthesis of three leaves per species was measured in one ambient and one elevated plot; measurements were rotated among the three blocks at approx. 1.5 h intervals. Measurements were made using a portable gas exchange system (Li-Cor 6400, Li-Cor Inc.) with a clear chamber head to allow for natural sunlight to illuminate the leaf. Measurements were made at the respective growth [CO2] of 370 µmol mol−1 or 550 µmol mol−1. Measurements were confined to the youngest fully expanded leaf of the leader branch as for the measurements of A vs Ci above. These leaves constituted the bulk of the plant canopy surface. The leaves were measured in a horizontal position to minimize within leaf variation and the effect of leaf angle on incident photon flux (Garcia et al., 1998). To ensure steady-state photosynthesis measurements while avoiding significant increases in leaf temperature, a stabilization period of at least 40 s was used before measurements (Long et al., 1996). Dew was observed on the leaves most mornings; visible moisture was removed with tissues before measurement. Mid-day values of stomatal conductance were analyzed to determine whether elevated [CO2] resulted in any differences in situ. Daily integrated rates of photosynthesis, A′, were estimated from the diurnal measurements of photosynthesis by fitting a fourth order polynomial to the data collected. The area under the fourth order curve was estimated in 15-min increments and summed for the whole day.
To determine whether changes in Vc,max and Jmax could predict responses of assimilation over the diurnal course, values of Vc,max and Jmax as estimated from the A vs Ci curves were used to predict diurnal rates of photosynthesis during July, 2000. This date was chosen because this was when diurnal measurements and A vs Ci measurements coincided most closely. For each measured diurnal time period, modeled values of A were estimated using the leaf model of photosynthesis (Farquhar et al., 1980) parameterized using the equations for Rubisco-limited A from Bernacchi et al. (2001) and for RuBP regeneration-limited A from Bernacchi (2003). The measured and modeled values of A were then compared by coplots of their diurnal variation.
Results of Asat, gs, Vc,max and Jmax determinations for control and elevated [CO2] grown leaves were analyzed using a complete block repeated measures anova using species, time and [CO2] treatment as main effects (Mixed Procedure, The SAS System 8.1; SAS Institute, Cary, NC, USA). Pre-determined contrasts between control and elevated [CO2] were performed on each species separately, regardless whether species by treatment interactions were shown to be statistically different between the two treatments. Measurements made, after the coppice, in June 2002 were analyzed separately from the precoppice measurements using a complete block ANOVA with species and treatment as main effects.
A comparison of regressions technique was used to test differences in diurnal rates of photosynthesis among the ambient and elevated grown [CO2] treatments (Mead & Curnow, 1983). Each day yielded only one value of A′ for each of the three species in the control and elevated [CO2] treatments. Therefore, the interactions between date of measurement and species were not possible to test statistically. The comparison of regressions technique required that the diurnal plots were fit to a polynomial that accurately reflects the diurnal changes in photosynthesis. A 4th order polynomial was shown to fit this criteria and was fitted first to both treatments together and then to each treatment separately. The 4th order accounted for significantly more variation than lower order polynomials, while 5th and higher order polynomials did not account for significantly more variation than the 4th order. The goodness-of-fit for each separate curve was compared with the goodness-of-fit for one line representing all data. The type 3 sum of squares (i.e. error variances around each model) and associated degrees of freedom for lumped vs separate fits were used to compute F-ratios associated with tests of homogeneity of the fitted polynomials (Mead & Curnow, 1983; Potvin et al., 1990). Significance values were set a priori at P < 0.1; this value is justified by the low level of replication, that is, three plots per treatment.
Growing three poplar species under field conditions in an atmosphere with [CO2] predicted for 2050 resulted in increased light-saturated rates of leaf photosynthesis (Asat; Fig. 1, Table 1). Statistical analysis of the data shows Asat to be highly variable with time and a significant time by treatment interaction was observed (Table 1). The statistically significant time by treatment interaction is likely caused by the relatively lower stimulation in Asat with elevated [CO2] during the initial measurements in the 2001 growing season. Predetermined treatment contrasts were performed on each species separately and all show statistically significant increases in Asat with growth in elevated CO2 (P < 0.001 for P. alba and P. nigra, P < 0.005 for P. × euramericana). Results for the maximum quantum efficiency of carbon assimilation () are similar to the results observed for Asat, with the only exception being that there was no treatment by measurement date interaction observed (Fig. 2, Table 1). Growth in elevated [CO2] only altered leaf level stomatal conductance for the species by measurement date interactions; there was no observed effect of [CO2] whether extracted from the photosynthetic light response curves measured on detached leaves or from mid-day measurement of attached leaves (Fig. 3, Table 1).
Table 1. Repeated measures analysis of variance of Vc,max, Jmax, , Asat, and gs for two species and one hybrid of poplar grown in control (370 µmol mol−1) and elevated (550 µmol mol−1) CO2 concentrations in FACE with species, date and treatment as main effects. There were three replicate plots per treatment for the poplar species with a minimum of two subsamples per plot
Species × treatment
Species × Date
Treatment × Date
Species × Treatment × Date
Summarizing the results from A vs Ci curves for the duration of the experiment, a statistically significant decrease in Vc,max was observed for plants grown in elevated CO2 (Table 1, Fig. 4) though individual contrasts show that only P. nigra demonstrated a decrease in Vc,max(P < 0.01) with growth in elevated [CO2]. Measured values of Vc,max were shown to change for all three species over measurement date (Fig. 4, Table 1). A statistically significant decrease in Jmax with growth in elevated [CO2] was also observed for the duration of the experiment (Fig. 5; Table 1), however, individual comparisons show that only P. alba (P < 0.03) and P. nigra (P < 0.04) resulted in a down-regulation of Jmax. Over the duration of the experiment, Jmax also varied with measurement date for all species.
To assess the in situ responses of growth in elevated [CO2], diurnal patterns of photosynthesis were measured at three points during the 2000 growing season. For each of these three dates, patterns of Q, vapor pressure deficit (VPD), and leaf temperature (Tleaf) were measured and are presented in Fig. 6. Growth and measurement in elevated [CO2] resulted in higher rates of photosynthesis for all three species over most of the diurnal course on each day (Fig. 6, Table 2). Diurnal photosynthesis measured in situ was closely predicted from Vc,max and Jmax measured on detached leaves from the same population. The only exception was in the late afternoon for P. alba when the modeled value over-estimates the in situ value (Fig. 6).
Table 2. The daily integral of net carbon assimilation (A′) for two species and one hybrid of poplar for three measurement days. Values were determined by estimating the area under a fourth order polynomial fitted to diurnal measurements of photosynthesis. P-values are for comparison of regressions technique on diurnal photosynthesis measurements (Mead & Curnow, 1983, see Materials and Methods for details)
Control mol m−2
Elevated CO2 mol m−2
P. × euramericana
P. × euramericana
P. × euramericana
Integrated over the day, growth at elevated [CO2] resulted in an increase in daily carbon uptake for upper canopy leaves ranging from 35% to 86%, depending on species and measurement date (Table 2). While there were not enough degrees of freedom to statistically compare the effect of elevated [CO2] among the three species, there were some noticeable trends. P. × euramericana apparently showed the highest daily integrals of net CO2 uptake (A′) for both treatments, but the relative stimulation by elevated [CO2] was only 46%. P. alba showed the lowest values of A′, though the relative stimulation was the largest (Table 2).
Re-growth after coppicing shows strong evidence of photosynthetic down-regulation for P. nigra and P. × euramericana where photosynthesis for leaves grown and measured in elevated [CO2] are similar to leaves grown and measured in current ambient [CO2] (Fig. 7). Since elevated [CO2] only resulted in a statistically significant decrease in gs for P. nigra (F1,10 = 18.85, P < 0.002; Fig. 7) the responses of A likely resulted from down-regulation of Vc,max (F1,26 = 9.23, P < 0.006) and Jmax (F1,26 = 5.13, P < 0.04; Fig. 8). Pre-determined contrasts revealed that P. nigra had the strongest down-regulation response of both Vc,max and Jmax, while only Vc,max was observed to down-regulate for P. × euramericana and no down-regulation was apparent for P. alba (Fig. 8).
The results show that even fast growing trees grown without rooting volume restriction in the open may still show a down-regulation of photosynthetic potential at elevated [CO2]. Despite down-regulation for two of the species, all species showed higher rates of carbon uptake for the upper canopy leaves when grown under elevated [CO2]. Against expectation there was no significant effect of growth at elevated [CO2] on stomatal conductance. In sharp contrast to all measurements before coppicing, an increase in Asat with growth in elevated [CO2] was not apparent for P. × euramericana and P. nigra during re-growth after the first coppice. This is explained by the large decrease in photosynthetic potential in these two genotypes. The following will address the responses of photosynthesis, conductance, and acclimation for the first growth cycle (1999–2001) and then for the postcoppice re-growth (2002).
Elevated [CO2] stimulated photosynthesis for all three species of poplar for the entire first growing cycle. Reviews of tree responses to elevated [CO2] show a mean increase in Asat ranging from 44 to 63%, though the response is highly variable (Gunderson & Wullschleger, 1994; Norby et al., 1999). These reviews, however, are heavily weighted to studies on isolated seedlings and saplings grown in artificial conditions with CO2 at or above 700 µmol mol−1 compared with 550 µmol mol−1 used in this FACE study. The mean response for the three poplars presented in this study is an approximate 38% increase in Asat measured at 25°C (Fig. 1). This increase in Asat for a 49% increase in [CO2] is likely close to the maximum theoretically possible without up-regulation of photosynthetic capacity (Long, 1991). Additionally, elevated [CO2] resulted in significantly higher values of for all three species (Fig. 2); this is commonly predicted to occur for plants growing under [CO2] enrichment (Long & Drake, 1991; Osborne et al., 1998). This increase in is attributed to a decrease in photorespiration as the availability of CO2 at the Rubisco active site increases (Drake et al., 1997).
There was significant photosynthetic down-regulation for P. nigra, Vc,max and Jmax, and for P. alba, Jmax (Figs 4 and 5). These decreases in photosynthetic potential were not sufficient to offset the stimulation of A by elevated [CO2]. Thus Asat for leaves grown and measured at elevated [CO2] always exceeded those of leaves grown and measured at ambient [CO2] (Fig. 1). Previous studies on woody species exposed to elevated [CO2] using growth-chambers or glasshouses have demonstrated more significant down-regulation responses (Pettersson & McDonald, 1992; Gunderson & Wullschleger, 1994; Sage, 1994; Loats & Rebbeck, 1999; Norby et al., 1999). One of the few other studies that address long-term tree responses to FACE did not show any evidence of photosynthetic down-regulation after 3 yr of growth in elevated [CO2] (Liquidambar styraciflua; Herrick & Thomas, 2001). These trees, however, were grown first at ambient [CO2] and then subjected to a step increase in [CO2] whereas at PopFACE, the trees were grown from planting to canopy closure in elevated [CO2].
Despite the 38% increase in Asat associated with growth in elevated [CO2], daily integrated rates of in situ photosynthesis were increased by 40 to almost 90% (Table 2, Fig. 6). This is explained by the fact that daytime leaf temperatures were typically over 30°C resulting in a larger stimulation of leaf photosynthesis by elevated [CO2] than would be evident at 25°C (Long, 1991). All species showed a large increase in daily integrated photosynthesis despite the photosynthetic down-regulation responses observed for P. nigra and P. alba. Because there were no observable changes in LAI (Gielen et al., 2001), it is likely that the increase in A′ resulted in higher rates of carbon assimilation at the canopy level for all three species. The stimulation in photosynthesis with growth under elevated [CO2] is consistent with the large increases in above- and below-ground biomass for these three species at PopFACE (Calfapietra et al., 2003).
The strong agreement between the modeled and measured rates of A throughout the diurnal time course shows that Vc,max and Jmax closely predict A under field conditions. P. alba, however, shows a period in the late afternoon when the modeled and measured values do not agree. Specifically, the modeled values suggest that rates should be higher than measured in the field for the elevated [CO2] treatment. Higher rates of photosynthesis associated with the elevated [CO2] would likely result in the accumulation of triose phosphates in leaves compared with the control treatments (Rogers et al., 1998; Isopp et al., 2000) and could lead to TPU-limitation of photosynthesis (Sharkey et al., 1986).
Despite only modest down-regulation responses of photosynthesis during the first rotation cycle, the spring following the harvest resulted in large down-regulation responses for two species such that elevated [CO2] did not stimulate rates of carbon assimilation compared with the control (Fig. 7). Of the three species, P. nigra shows the strongest evidence of down-regulation for both Vc,max and Jmax while P. × euramericana showed a down-regulation response for only Vc,max (Fig. 8). Coupled with our precoppice measurements (Figs 4 and 5) these results suggest that P. nigra may be more sink limited than the other two species, i.e. it has less capacity to utilize the additional photosynthate generated under elevated CO2. This also corresponds with the observation that P. nigra is the only species to show a significant decrease in gs (Fig. 7). It is likely that the decrease in gs is the result, rather than a cause, of the loss of photosynthetic potential, as has been demonstrated previously for soybean (Fiscus et al., 1997) and for other poplars (Noormets et al., 2001).
As a result of the down-regulation responses, the enhancement of net photosynthesis by decreased photorespiration under elevated [CO2] is offset by decreased photosynthetic potential during the initial stages of re-growth. Coppice re-growth depends on remobilization of reserves from the vascular parenchyma of the root system. Calfapietra et al. (2003) have shown that root biomass of these trees was increased by 20–30% under elevated [CO2]. If the coppice developed in elevated [CO2] contains a higher reserve of carbohydrate, then it is possible that more soluble carbohydrates are translocated to the regrowing shoot and that this causes a down-regulation of photosynthetic potential for P. nigra and P. ×euramericana. Increased soluble carbohydrate concentrations in the leaf have been linked to suppression of some of the genes coding for the photosynthetic apparatus, in particular rbcS which codes for the small subunit of Rubisco (Sheen, 1990; Drake et al., 1997; Moore et al., 1998, 1999). This may explain the photosynthetic down-regulation observed for P. nigra and P. × euramericana grown in the elevated [CO2] treatment during re-growth compared with the first rotation cycle when the roots and shoots were developing simultaneously. P. alba, however, did not demonstrate a down-regulation in photosynthetic potential, though root biomass was also higher in elevated [CO2] for this species (Calfapietra et al., 2003).
This research was conducted within the framework of the EU-POPFACE (ENV4-CT97-0657) program. Research was partially funded by the Integrative Photosynthesis Research grant to CJB (NSF DBI96-02249) and by the Special Undergraduate Research on the Environment (SURE) from the University of Illinois Environmental Council to VEW. The authors acknowledge Patrick Morgan, Shawna Naidu, David Moore, Emily Heaton, and Andrew Leakey for comments on this manuscript, Franco Miglietta for design and maintenance of the PopFACE Facility, and Beverly Bryant for assisting with measurements. We further acknowledge Guy Bernacchi and Best Travel and Tours, Inc. of Chicago, IL USA for supplying airfare.