INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The majority of experimental evidence points to a stimulation of light-saturated photosynthesis (A_sat) for C₃ plants, grown in the atmospheric partial pressure of CO₂ (pCO₂) predicted for the end of this century (for review, see Drake et al., 1997). In the field, increased photosynthesis in elevated pCO₂ has been shown to both increase and decrease photochemical requirements for light-saturated electron flow through photosystem (PS) II (J_t; Scarascia-Mugnozza et al., 1996; Hymus et al., 1999). What basis might there be for a variable response in electron transport when assimilation is consistently increased?

Elevated pCO₂ will stimulate the photosynthetic carbon reduction cycle and the electron flow that drives it (J_c). However, elevated pCO₂ will also competitively suppress the photorespiratory carbon oxidation (PCO) cycle and the electron flow that drives it (J_o). Whether or not there is an increase in the demand of carbon metabolism for J_t will depend on the net effect of these changes in J_c and J_o. The mechanistic understanding of C₃ photosynthesis proposed by Farquhar et al. (1980) predicts that, when pCO₂ is increased, if A_sat is limited by the amount of Rubisco the stimulation of J_c will be greater than the suppression of J_o, and an increase in J_t will result. This predicted increase in J_t may not be observed where growth in elevated pCO₂ results in acclimation of either: (a) the amount of Rubisco in the leaf, or (b) sinks for J_t other than the photosynthetic carbon reduction and PCO cycles. Decreased leaf Rubisco content in elevated pCO₂ will decrease both J_c and J_o. In addition, J_o will be competitively suppressed by increasing pCO₂. Many studies have shown no effect of growth in elevated pCO₂ on sinks for J_t, other than carbon metabolism (Epron et al., 1994; Habash et al., 1995; Bartak et al., 1996; Hymus et al., 1999). However, there is some evidence that growth in elevated pCO₂ decreases antioxidant activity (Polle et al., 1997), suggesting a possible change in potential electron flux to a Mehler reaction.

Where elevated pCO₂ changes photochemical quenching of absorbed photosynthetically active photon flux density (PPFD), changes in non-photochemical quenching of absorbed PPFD will result. Non-photochemical quenching constitutes a dynamic form of photoinhibition (Baker and Ort, 1992; Long et al., 1994; Osmond, 1994). At high light, if elevated pCO₂ increases photochemistry, a decrease in non-photochemical quenching and protection against photoinhibiton would be expected.

In this study, Dactylis glomerata was grown under controlled environment conditions, at two levels of nitrogen nutrition. A previous study of D. glomerata showed decreased leaf Rubisco content with growth in elevated pCO₂, but only when nitrogen supply was limiting (Davey, 1998). Given this potential to change leaf Rubisco content of plants growing in elevated pCO₂ in controlled environments, the following two hypotheses were tested: (a) In the absence of acclimation, elevated pCO₂ will result in a net increase in J_t because the stimulation of J_c will be greater than the inhibition of J_o, decreasing photoinhibition; and (b) an acclimatory decrease in leaf Rubisco content in elevated pCO₂ will offset the stimulation of J_c, J_o will be suppressed, J_t will decrease, and photoinhibition will increase in elevated pCO₂.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Light-Saturated Photosynthesis

Growth in elevated pCO₂ did not affect V_c,max or J_max in the high-nitrogen treatment (Fig. 1a; Table I), where V_c,max is the maximal Rubisco catalyzed carboxylation rate and J_max the maximal whole-chain electron transport rate. In the low-nitrogen treatment, V_c,max was significantly decreased, by 42%, in elevated pCO₂, the reduction in J_max was not significant (Fig. 2a; Table I). In high nitrogen, A_sat was Rubisco limited under ambient pCO₂. As a consequence, A_sat, J_c, and J_t were significantly increased, by 86%, 73%, and 55%, respectively, for plants grown and measured at elevated pCO₂, relative to those grown and measured at ambient pCO₂ (Fig. 1a; Table I). In the low-nitrogen treatment, acclimation to elevated pCO₂ resulted in no stimulation of A_sat and a significant decrease in both J_c and J_t of 11% and 20%, respectively, when the comparisons at respective growth pCO₂ were made (Fig. 2a; Table I). In high nitrogen, the suppression of photorespiration by elevated pCO₂ reduced J_o by an apparent 16%. In low nitrogen, J_o was decreased by 57% due to suppression of photorespiration by elevated pCO₂ and decreased Rubisco (Table I). Neither decrease in J_o was statistically significant.

View larger version (22K): [in this window] [in a new window]   Figure 1.   Light-saturated photosynthesis: high nitrogen. a, The responses of light-saturated CO2 uptake (Asat) against intercellular CO2 concentration (Ci) for leaves of D. glomerata grown in high nitrogen and at either elevated (black symbols and black lines) or current ambient (white symbols and dotted lines) pCO2. Values of Vc,max and Jmax, calculated using the equations and constants in von Caemmerer (2000) and Bernacchi et al. (2001), were used to fit a nonlinear regression to observed values above (Jmax) and below (Vc, max) the inflection of the curves. Also shown are the supply functions for each curve (dashed line) that indicate the operating point of photosynthesis at the growth pCO2 for each treatment. Data points shown are the means (±1 SE) for five replicate leaves. Measurements were made in 21 kPa O2 and at a PPFD of 1,300 µmol m2 s1. b, Jt, Jc, and Jo for ambient (white bar) and elevated (black bar) pCO2 treatments were calculated for measurements at the respective growth pCO2 for each group of leaves using the equations of Valentini et al. (1995). Values shown are the means (±1 SE) for five replicate leaves.

View this table: [in this window] [in a new window]   Table I.   Light-saturated photosynthetic characteristics Asat, the quantum efficiency of PSII (PSII), photochemical quenching coefficient (qP), and F'v/F'm measured simultaneously at the growth pCO2 for ambient (35 Pa) and elevated (65 Pa) treatments. Vc,max and Jmax were calculated using the equations and constants in von Caemmerer (2000) and Bernacchi et al. (2001). Jt, Jc, and Jo were estimated using the model of Valentini et al. (1995). All values are the means (±1 SE) for five replicate plants. A significant interaction between growth pCO2 and nitrogen was found for each parameter measured, except Jo. As a consequence, differences between individual means were tested using a post hoc Tukey's test. Different superscript letters identify means that are significantly different within that row (P < 0.05). Asat, Vc,max, Jmax, Jt, Jc, and Jo are expressed in µmol m2 s1. Units for PSII, qP, and F'v/F'm are dimensionless.

View larger version (21K): [in this window] [in a new window]   Figure 2.   Light-saturated photosynthesis: low nitrogen. a, Plot of light-saturated A against Ci for leaves of D. glomerata grown in low nitrogen. As described previously for Figure 1. b, Measurements of Jt, Jc, and Jo made at the respective growth pCO2 for D. glomerata grown in low nitrogen. As described previously for Figure 1.

In the high-nitrogen treatment, elevated pCO₂ significantly increased _PSII when measurements made at the two growth pCO₂ were compared, as a result of increases in both F_v'/F_m' and photochemical quenching coefficient (q_P; Fig. 1b; Table I), where F_v'/F_m' is the probability of an absorbed photon reaching an open PSII reaction center. In low nitrogen, F_v'/F_m' was significantly lower in elevated pCO₂; the decreases in _PSII and q_P were not statistically significant (Fig. 2b; Table I).

Ratio of Electron Transport to CO₂ Fixation

For all nitrogen and pCO₂ treatments, _PSII and _CO2 were highly correlated (r² = 0.74-0.82; P < 0.05) and linearly related with an intercept that was not significantly different from zero (Table II), where _CO2 and _PSII are the quantum efficiencies of CO₂ uptake and of linear electron transport through PSII, respectively. It is important that growth pCO₂ had no statistically significant effect on the _PSII/_CO2 relationship for plants grown in either high or low nitrogen (Table II). These relationships indicated that growth in elevated pCO₂ had not resulted in additional sinks for J_t, such as to a Mehler reaction.

View this table: [in this window] [in a new window]   Table II.   Ratio of electron transport to CO2 fixation The gradient (k) of the relationship PSII/CO2 was unaffected by elevated pCO2 in low nitrogen (F1.64 = 2.1; P > 0.05) and high nitrogen (F1.62 = 2.79; P > 0.05). The intercept (b) was not significantly different from zero for each plot. Low nitrogen ambient, t1.29 = 0.02, P = 0.98; low nitrogen elevated, t1.31 = 0.95, P = 0.34; high nitrogen ambient, t1.34 = 1.14, P = 0.26; high nitrogen elevated, t1.28 = 0.34, P = 0.73.

Photoinhibition and Recovery

F_v/F_m measured prior to the beginning of the photoperiod was unaffected by growth pCO₂ for both nitrogen treatments (F_1,16 = 0.1; P > 0.1). For all treatments, F_v/F_m declined during the 3-h high-light treatment at a PPFD of 2,000 µmol m² s¹ (Figs. 3a and 4a). After 3 h, F_v/F_m was significantly higher by 7% in elevated pCO₂ for the high-nitrogen plants (t₁₈ = 2.3; P < 0.05; Fig. 4a). In low nitrogen, F_v/F_m measured after 3 h was unaffected by pCO₂ (Fig. 4a).

View larger version (22K): [in this window] [in a new window]   Figure 3.   Photoinhibition and recovery: high nitrogen. a, Photo-inhibitory reduction in Fv/Fm. b, Changes in Fo during a 3-h exposure to a PPFD of 2,000 µmol m2 s1. c, Recovery of Fv/Fm measured after 10 min dark adaption (black lines), and Fv'/Fm' measured under growth PPFD (dashed lines), for D. glomerata grown in high nitrogen. Plants were grown, photoinhibited, then allowed to recover in their growth pCO2 either ambient (white symbols) or elevated (black symbols). Each symbol represents the mean (±1 SE) for 10 replicate plants.

View larger version (20K): [in this window] [in a new window]   Figure 4.   Photoinhibition and recovery: low nitrogen. a, Photo-inhibitory reduction in Fv/Fm. b, Changes in Fo during a 3-h exposure to a PPFD of 2,000 µmol m2 s1. c, Recovery of Fv/Fm measured after 10 min of dark adaption (black line), and Fv'/Fm' measured under growth PPFD (dashed line), for D. glomerata grown in low nitrogen. As described previously for Figure 3.

In high nitrogen with elevated pCO₂, 60% of the reduction in F_v/F_m had recovered after 10 min of dark adaptation and after 1 h, F_v/F_m had returned to dark-adapted levels. Although the ambient pCO₂ treatment similarly recovered after 1 h, F_v/F_m and F_v'/F_m' after 10 min were lower than in elevated pCO₂. This difference in F_v/F_m was statistically significant (t₁₈ = 3.5; P < 0.05; Fig. 3c).

For plants grown in low nitrogen, the recovery of F_v/F_m took between 3 and 4 h (Fig. 4c). Values of F_o measured before and at the end of the 3-h treatment were not significantly different in either the ambient (t₁₈ = 1.5; P > 0.1) or elevated (t₁₈ = 0.01; P > 0.1) pCO₂ treatments (Fig. 4b). Although there was no effect of pCO₂ on the recovery of F_v'/F_m' in the low-nitrogen treatment, the initial recovery of F_v/F_m was significantly reduced by elevated pCO₂ during the first (t₁₈ = 3.9; P < 0.05) and second (t₁₈ = 3.0; P < 0.05) hours of the recovery. Given that F_v'/F_m' was unaffected by pCO₂, the slower recovery of F_v/F_m should reflect an affect of pCO₂ on the dark-adapted relaxation of non-photochemical quenching.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

This controlled-environment study confirmed the hypothesis that elevated pCO₂ decreases photoinhibition in high nitrogen, but increases photoinhibition in low nitrogen, at a level assumed to restrict growth. This is explained by a greater demand for electrons in photosynthetic carbon metabolism in the absence of limitations, and a decreased demand by both photosynthetic and photorespiratory carbon metabolism, relative to ambient pCO₂, when resources other than carbon restrict production.

For D. glomerata, light-saturated photosynthesis at ambient pCO₂ was limited by the amount of Rubisco, regardless of nitrogen treatment (Fig. 1a). At high nitrogen, elevated pCO₂ had no effect on V_c,max. Therefore, in keeping with the theory of Farquhar et al. (1980), elevated pCO₂ stimulated J_c to a greater extent than it suppressed J_o, increasing J_t (Fig. 1b; Table I). As a consequence, when exposed to a high PPFD, the proportion of absorbed photons dissipated photochemically was increased in the elevated pCO₂ treatment, reducing photoinhibition (Fig. 3).

For D. glomerata grown under low nitrogen supply, acclimation of the photosynthetic apparatus significantly reduced V_c,max (Fig. 2). The magnitude of the acclimation was sufficient to totally remove the short-term stimulation of J_c by elevated pCO₂ (Fig. 2). Because photorespiration was suppressed, a similar A_sat at elevated pCO₂ to that in ambient pCO₂, was achieved with about 20% lower J_t (Table I). If triose-phosphate utilization (TPU) had been limiting photosynthesis, we would have expected a similar decrease in J_t. Under conditions of TPU limitation, A_sat and J_c will be insensitive to increasing pCO₂ and J_o will be suppressed, and decreased J_t and increased non-photochemical quenching can result (Sharkey et al., 1988; Pammenter et al., 1993). In this study, A_sat of leaves grown in low nitrogen and elevated pCO₂ increased with an increase in C_i to 150 Pa. As a consequence, decreased V_c,max, not TPU limitation of photosynthesis, was responsible for decreased J_t. The nitrogen dependence of the pCO₂-dependent decrease in carboxylation capacity observed here is consistent with other studies of the interactive effects of growth at elevated pCO₂ and nitrogen supply (Tissue et al., 1993; Thomas et al., 1994; Curtis, 1996; Rogers et al., 1996). Rogers et al. (1998) showed, with a related herbage grass, Lolium perenne, that when nitrogen supply was limited there was a loss of carboxylation capacity and Rubisco at elevated pCO₂, but not when nitrogen supply was adequate. Partial defoliation removed this acclimatory response. This showed that low nitrogen affected acclimation by limiting sink size relative to source because when source size was decreased acclimation was removed. For D. glomerata, in low nitrogen, the deceased demand for photochemical energy was reflected in a significant depression in F_v'/F_m' in elevated pCO₂ and therefore increased the probability of absorbed photons being dissipated as radiation-less decay from the antenna of PS II (Table I, Fig. 2). The gradient of the straight line describing the dependence of _PSII on _CO2 was not significantly different in elevated pCO₂ for either nitrogen treatment. This suggested that growth in elevated pCO₂ did not produce significant sinks for photochemical energy other than the photosynthetic carbon reduction or PCO cycles, in agreement with published findings (Epron et al., 1994; Habash et al., 1995; Bartak et al., 1996; Hymus et al., 1999). However, there was no evidence of any significant sink beyond photosynthetic and photorespiratory carbon metabolism at the current ambient pCO₂ in D. glomerata. In other species, notably those that bare leaves throughout long periods of environmental stress restricting carbon metabolism, alternative sinks for electron flow, primarily to oxygen, are suggested to be very significant (Lovelock and Winter, 1996; Cheeseman et al., 1997). Responses of species with these stress tolerance strategies might be very different.

After the 3-h photoinhibitory treatment, the recovery of F_v/F_m in low nitrogen was slower for the elevated pCO₂ treatment. Because F_o was unaffected (Fig. 3), the decreased F_v/F_m was almost certainly associated with zeaxanthin-dependent quenching (Demmig-Adams and Adams, 1992; Owens, 1994; Horton et al., 1996). Under stress conditions, this recovery can require many hours (Demmig-Adams and Adams, 1992; Fryer et al., 1995). The results suggest that in addition to increasing the potential for photoinhibition, elevated pCO₂ may also decrease capacity for recovery. Hymus et al. (1999) showed that loblolly pine (Pinus taeda) photoinhibited during winter low temperature stress recovered more slowly when growing at elevated pCO₂. Together, these results indicate that elevated pCO₂, may decrease the capacity of the plant to recover from stress-induced photoinhibition. This may be part of a wider pattern of decreased stress tolerance in leaves growing at elevated pCO₂ (Lutze et al., 1998; Terry et al., 2001).

In a study on wheat (Triticum aestivum) grown for 6 weeks under optimal conditions, without an acclimatory loss of Rubisco, both A and total non-cyclic electron flow through PSII were enhanced by elevated pCO₂ at high light (Habash et al., 1995). A similar study on ryegrass (Lolium perenne) showed that A and _PSII were increased by instantaneous elevation of pCO₂ for plants grown at current pCO₂, but longer term acclimation completely reduced the stimulation of both A and _PSII (Bartak et al., 1996). For natural vegetation growing in the field, studies show seasonal decreases in photochemistry, and increased photoinhibition in elevated pCO₂ (Scarascia-Mugnozza et al., 1996; Hymus et al., 1999).

Here, we have extended these findings by providing quantitative evidence that elevated pCO₂ can either increase or decrease photochemistry and photoinhibition, depending on whether down-regulation of photosynthetic capacity has occurred. Plants were grown at 400 µmol m² s¹, but exposed to higher light for the photoinhibitory treatment. This is not unrealistic of many areas of the globe where a series of cloudy days may be followed by clear sky days with higher photon flux or where grazing exposes lower canopy leaves to high light. This finding has important ecological implications. Although under optimal conditions, elevated pCO₂ increases photochemical energy use and decreases the probability of photoinhibition, the reverse is true of limiting nitrogen conditions. Most of the natural terrestrial biosphere and much subsistence agriculture is nitrogen limited. It has been widely appreciated that the response of photosynthesis to rising pCO₂ may be diminished by acclimation in these conditions. Here, we show that not only is capacity for carbon assimilation decreased, but the probability of photoinhibition, due to increased non-photochemical quenching, is increased. Such non-photochemical quenching would serve to protect the reaction centers from photo-inactivation and damage when the rate of excitation of PSII is in excess of the rate of photochemistry. The cost of this protection is that when a leaf is in low light after photoinhibition the efficiency of photosynthesis remains low for minutes to hours, resulting in a significant loss of potential carbon fixation (Long et al., 1994). Elevated pCO₂ will increase this loss, both by increasing the potential for photoinhibition and by slowing the rate of recovery.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Growth Conditions

Dactylis glomerata (IGER, Aberystwyth, UK) was grown from seed for 56 to 57 d in a washed silver sand media (William Sinclair Horticulture, Lincoln, UK), in 0.62-L pots. Four seeds were sown in each pot. These were then divided between two controlled environments (PG660, Sanyo, Loughborough, UK); one was maintained at 35 Pa pCO₂ (ambient), the other at 65 Pa pCO₂ (elevated). An infrared gas analyzer integrated with a feedback control system (WMA-2, PP Systems, Hitchin, UK), which controlled the injection of scrubbed, pure CO₂ gas (Linde Gas Ltd, Stoke on Trent, UK), maintained the pCO₂ within the controlled environment cabinets. Plants were grown in a day/night temperature regime of 16°C/12°C and a relative humidity of 80%, giving a daytime water vapor pressure deficit of 0.5 kPa. The photoperiod was 14 h long at a PPFD of 400 µmol m² s¹ at pot height, providing a total photon flux of 20 mol m² d¹ over the photoperiod and similar to that which D. glomerata would receive in the field during summer in Western Europe.

From planting to 1 week after emergence, the sand media was fully saturated with deionized water. At this point, the plants were divided into two nutrient regimes in each cabinet. They were supplied with either high (12 mM) or low (4 mM) nitrogen by a Long Ashton (nitrate type) solution (Hewitt, 1966). Throughout the growth period, the sand media was flushed twice weekly with deionized water to prevent accumulation of salts.

To minimize undetected inter-cabinet environmental differences, pCO₂ treatments and their plants were swapped between cabinets each week. To minimize the effects of intra-cabinet environmental gradients, the plants were randomly repositioned within the cabinets each week. Between 56 and 57 d into growth, five plants of the eight grown for each treatment were randomly selected for simultaneous measurements of leaf gas exchange and chlorophyll a fluorescence. Measurements were made on the youngest leaf with an emerged ligule on the main stem.

Leaf Gas Exchange

Leaf net CO₂ uptake (A) and water vapor efflux were measured in an open gas exchange system. A combined CO₂ and water vapor analyzer (LI-6262, LI-COR, Lincoln, NE), calibrated against a water vapor generator (WD600, ADC Ltd., Hoddesdon UK) and a standard CO₂ concentration of 50 Pa (Linde Gas Ltd) was used. Inlet CO₂ concentration was controlled by a gas dilutor (GD-600, ADC Ltd.), and inlet humidity was controlled by passing the dry airflow through a temperature-controlled ferrous-sulfate crystal column (WG-600, ADC Ltd.). The temperature-controlled leaf section chamber used (LSC, ADC Ltd.) allowed for rapid mixing of gases, and a small response time to changes in pCO₂ and PPFD. Chamber cooling was by circulating coolant through each half of the chamber. All measurements were made at a leaf temperature of 16.0°C (±0.3°C) and a water vapor pressure deficit of 1.2 (±0.04) kPa.

The light-saturated response of A to C_i was made at a PPFD of 1,300 µmol m^{2 s1}. Photosynthetic induction was performed at the growth pCO₂. Calculations of A and C_i followed von Caemmerer and Farquhar (1981). V_c,max and J_max were estimated for each individual leaf by fitting maximum likelihood regressions to the initial slope and plateau of the A/C_i response curves, respectively, using the calculations of von Caemmerer (2000) and Bernacchi et al. (2001).

Leaf Chlorophyll a Fluorescence

A modulated chlorophyll fluorimeter and leaf clip (PAM 2000, H Walz, Effeltrich, Germany) were used to measure minimum (F_o'), maximum (F_m'), and steady-state (F_s) levels of fluorescence simultaneously with the gas exchange measurements. These values were used to calculate the efficiency of excitation energy capture by open PSII reaction centers (F_v'/F_m'), the q_P, and _PSII (Genty et al., 1989).

The response of A and _PSII to PPFD was determined over a range of light levels from 0 to 1,420 µmol m² s¹ under non-photorespiratory conditions (1 kPa O₂; Linde Gas Ltd). Leaf absorptance () was measured with a Taylor integrating sphere attached to a quantum sensor (SKP 215, Skye Instruments Ltd, Llandrindod Wells, UK) following the method of Rackham and Wilson (1968). From these measurements, _CO2 was then calculated as:
where Q is PPFD. When measured in 1% (v/v) O₂, the relationship of _PSII to _CO2 is linear (Genty et al., 1989). The model and assumptions of Valentini et al. (1995) use the _PSII/_CO2 relationship in 1% (v/v) O₂, to calculate light-saturated linear electron flow through PSII (J_t) and partition it between electron flow to the photosynthetic carbon reduction (J_c) and PCO (J_o) cycles. In this model, the linear relationship between _PSII and _CO2 is assumed to describe the apparent quantum efficiency of photosynthetic linear electron flow (_e) using the equation:
where 4 is the number of electrons needed per CO₂ molecule fixed, k is the gradient, and b is the y intercept. The model assumes that this relationship holds in both photorespiratory and non-photorespiratory conditions, enabling J_t to be calculated as:
By assuming only the photosynthetic carbon reduction and PCO cycles are sinks for linear electron flow, the model calculates the partitioning of J_t between J_c and J_o as:


where 4 is the number of electrons required to fix one molecule of CO₂ and R_l is the rate of CO₂ production by photorespiration. J_c and J_o are then solved as:

Photoinhibitory Treatment

Between 54 and 58 d into growth, leaves of D. glomerata grown in the nitrogen and pCO₂ treatments described previously were selected for a photoinhibitory, high-light treatment. The criteria used for leaf selection were as for the previous measurements. Leaves were exposed to a PPFD of 2,000 µmol m² s¹ for 3 h, within a custom-built controlled environment in which the pCO₂ was maintained at growth levels, air temperature was maintained at 16°C to 18°C, and relative humidity was maintained at 75% to 82%, approximating growth conditions. Eight plants, two from each treatment, were randomly selected and photoinhibited on each of 5 consecutive d. In total, 10 plants from each treatment were photoinhibited.

Measurements of F_o and F_m were made in the dark prior to the beginning of the photoperiod to determine the maximum quantum yield of PSII (F_v/F_m), then every 45 min during the photoinhibitory treatment, using a modulated chlorophyll fluorometer (PAM 2000). The photoinhibitory treatment was begun 1 h into the photoperiod. After 3 h, the plants were returned to their respective growth environments to recover. The recovery of F_v/F_m measured after 10 min of dark adaptation, and F_v'/F_m' measured under growth light levels, was followed. Measurements were made immediately following the photoinhibitory treatment and then at hourly intervals until the recovery was complete, using the equipment and protocols for measurement described previously.

Statistical Analysis

The effects of growth pCO₂ and nitrogen supply on A_sat, V_c,max, J_max, _PSII, q_P, F_v'/F_m', J_t, J_c, J_o, and  were tested using ANOVA. Where a significant interaction between pCO₂ and nitrogen was found, post hoc pair-wise comparisons using Tukey's test were performed to identify differences between individual means. The effect of pCO₂ treatment on the recovery of F_v/F_m was tested using a two-tailed Student's t test. All ANOVA and Student's t tests were performed using statistical software (Systat 7.0, Systat Inc, Evanston, IL). The effect of pCO₂ treatment on the relationship between _PSII and _CO2 was examined by regression analysis of variance (Zar, 1999). Fluorescence yields were arcsine transformed to generate a normal distribution for statistical analysis (Zar, 1999). An effect was described as significant where P < 0.05.