Environmental Pollution 109 (2000) 501±507
Beech (Fagus sylvatica) response to ozone exposure assessed with a chlorophyll a ¯uorescence performance index A.J. Clark a,b, W. Landolt a, J.B. Bucher a, R.J. Strasser b,* a
Swiss Federal Institute for Forest, Snow and Landscape Research, CH-8903 Birmensdorf, Switzerland b Bioenergetics Laboratory, University of Geneva, CH-1254 Jussy Geneva, Switzerland Received 30 June 1999; accepted 5 January 2000
``Capsule'': A chlorophyll a performance index obtained from ¯uorescence analysis correlates to visual injury and biomass loss caused by ozone, and thus could be used for the large-scale monitoring of tree vitality. Abstract This paper describes a relationship between ozone exposure, biomass, visual symptoms and a chlorophyll a ¯uorescence performance index for young beech trees (Fagus sylvatica). The plants were exposed to four levels of ozone in open-top fumigation chambers (50, 85, 100% of ambient, and 50% of ambient+30 nl lÿ1 ozone) that ¯uctuated in parallel with ambient ozone during a single growing season. The trees were fumigated in the four treatments with ozone levels corresponding to an AOT40 (accumulated exposure above a threshold of 40 nl lÿ1) of 0.01, 3.35, 7.06 and 19.70 ml lÿ1 h, respectively. Highly signi®cant dierences were found between the 50% of ambient+30 nl lÿ1 ozone treatment and all other treatments, with a 70.5% reduction in primary photosynthetic performance, as measured with the PI index. The reduction of the PI values demonstrated a high correlation with visual symptom development (r2=0.98), and by the end of September with biomass loss (r2=0.99). A signi®cant ozone exposure±response relationship was found between AOT40 and primary photochemistry (r2=0.97). Thus, analysis of PI provides an alternative method for regional monitoring of tree health within the context of the currently employed AOT40. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Fluorescence; Ozone; Biomass; Performance index; Visual symptoms
1. Introduction Current ozone levels remain a risk to European forests (SkaÈrby et al., 1998). In the determination of an appropriate ozone critical level, a two-tier approach was suggested at the 1996 workshop under the UN/ECE convention on Long-Range Transboundary Air Pollution for abatement strategies to protect vegetation from ozone injuries. The Level I approach prescribed biomass loss to be the appropriate biological response parameter and that beech would be the most appropriate species to monitor (KaÈrenlampi and SkaÈrby, 1996; Fuhrer et al., 1997). The Level II approach proposed several additional environmental parameters to also be included due to their modifying in¯uence on tree ozone responses. It was also stated that there was a scarcity of data upon * Corresponding author. Tel.: +41-22-7591944; fax: +41-227591945. E-mail address: [email protected]
which to base an ozone dose±response function, especially for Level II. An important goal is the mapping of a physiologically eective ozone concentration in order to monitor and protect the European forests. An objective and simple screening method that is sensitive and suitable for both Level I and Level II assessments is required. Currently, visual identi®cation of ozone damage to leaves and needles is the only simple method of diagnosis, which often requires specialised knowledge because the environment and other stresses can modify the patterns of ozone injury. The lack of suitable monitoring techniques makes regional mapping of actual ozone-induced forest stress virtually impossible. A promising approach could be the use of chlorophyll (Chl) a ¯uorimetry which can provide large amounts of accurate data with a minimum of expertise and time and without injury to the plants. The techniques could be cost eective, objective, and the mapping of plant physiological condition a possibility.
0269-7491/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S0269-7491(00)00053-1
A.J. Clark et al. / Environmental Pollution 109 (2000) 501±507
Photosynthesis is a core function in the physiology of all plants. Its functional state has been considered an ideal physiological activity to monitor when the health and vitality of plants is under scrutiny. Measurements of photosynthesis have often been used in the assessment of ozone injury (Heath, 1996), and although the actual mechanisms by which ozone exposure alters photosynthesis are not yet known, it is a sensitive diagnostic parameter (Saxe, 1991). The oxidative byproducts from reactions occurring at cell membranes are believed to restrict carbon assimilation (Dann and Pell, 1989; Reddy et al., 1993; Heath, 1996). Consequently, a change in the energy capture process around photosystem II (PSII) and its concurrent ¯uorescence emissions occur (Farage et al., 1991; Mikkelsen, 1995). It is unlikely that PSII is directly aected by ozone, except possibly at very high concentrations. Chl a ¯uorescence emissions are, however, known to be sensitive to many of the environmental parameters chosen in the Level II approach as well as ozone (Lichtenthaler and Rinderle, 1988; Larcher, 1995; Mikkelsen, 1995; Strasser and Strasser, 1995). The aim of this paper was to test the sensitivity of a newly described Chl a ¯uorescence performance index for plant screening (Strasser et al., 1999), and to analyse the response of the performance index in relation to biomass and visual symptom development for beech trees growing in open-top chambers (OTCs). 2. Materials and methods 2.1. Plant material and ozone treatments The experiment was performed in an OTC facility at the Swiss Federal Institute for Forest, Snow and Landscaped Research (WSL), Birmensdorf, Switzerland. A total of 20 chambers were divided between four ozone treatments that were dependent upon prevalent ambient levels of ozone in the 1997 growing season between the beginning of April and the end of September (Table 1). Ozone was produced by electrical discharge from pure oxygen by an ozone generator (Model 502, Fischer, Germany), and concentrations were measured with an
ozone monitor (Model 8810, Monitor Labs, USA). Ozone concentrations within the chambers were regulated to oscillate in parallel with ambient ozone concentrations. At high ambient light levels (in excess of 600 mmol mÿ2 sÿ1) partial shading was automatically provided for the plants with the use of a translucent shading canopy which covered all chambers to maintain the chamber temperatures within an optimal range. Fagus sylvatica from two Swiss provenances, Hirschthal and Aarburg, were used in the experiment. All saplings were potted in a standard soil mixture (50% sandy soil, 25% brown earth, 25% peat, 2 g lÿ1 fertiliser (Osmocote plus) that was kept moist throughout the experiment. There was a single sapling from each provenance in each chamber, with ®ve replicate chambers within each treatment. Visual symptom development was assessed each week from two leaves per tree per chamber in accordance with GuÈnthardt-Goerg et al. (1993). The biomass data were calculated from sets of approximately 150 seedlings grown in the same OTCs, as described by Landolt et al. (2000). Parallel ¯uorescence measurements were taken from both the seedlings and saplings. 2.2. Chl a ¯uorescence measurements Chl a ¯uorescence transients were measured at ambient temperatures within the chambers using a portable ¯uorimeter (Model PEA, Hansatech Inst., UK). All fast ¯uorescence transients were recorded up to 2 s with 12bit resolution and a data acquisition rate of 10 ms in the time span from 10 ms to 2 ms, while after 2 ms and 1 s the instrument automatically switches to slower digitisation rates (for details see Strasser et al., 1995). The ¯uorescence emission was induced by an homogenous illumination on a 4-mm-diameter area of the leaf samples by red light (peak at 650 nm) of 3000 mmol mÿ2 sÿ1 provided by an array of six light-emitting diodes. Chl a ¯uorescence transients (1600) were recorded between 23:00 and 03:00 from leaves that had been in the dark for a minimum of 3 h to attain a fully dark-adapted state of all samples. Four primary ¯uorescence values parameters were retained. The ¯uorescence intensities at 50, 300 ms, and 2 ms, were denoted as F50 ms, F300 ms, and
Table 1 The four ozone exposure treatments (I±IV) used in the fumigation of beech trees in open-top chambers, their derivations in relation to the ambient level of ozone, and measured ozone concentrations and cumulative ozone exposure (AOT40) from April to September 1997 Treatment
I II III IV a
Pre-industrial level Reduced level Level at low altitudes (400±700 m a.s.l.) Level at high altitudes (1600±1800 m a.s.l.) S.D., 1 standard deviation.
Ozone level relative to ambient air
Ozone concentration (nl lÿ1) Mean
50% 85% 100% 50%+30 nl lÿ1
15.3 23.5 27.4 44.0
8.2 12.6 15.1 13.0
0.5 1.0 0.9 3.3
48.7 79.9 102.6 93.5
AOT40 (ml lÿ1 h) 0.01 3.35 7.06 19.70
A.J. Clark et al. / Environmental Pollution 109 (2000) 501±507
F2 ms respectively, and FM is the maximum ¯uorescence intensity. These values were used to calculate the performance index, denoted as PI (Clark et al., 1999; Strasser et al., 1999) which has been de®ned as the ratio of two recently described Structure±Function-Indexes (SFI). The ®rst, SFIP (Tsimilli-Michael et al., 1998) responds to structural and functional PSII events leading to electron transport within photosynthesis. The second, SFIN (Strasser et al., 1999), refers to the energy that is dissipated or lost from photosynthetic electron transport. It has been de®ned as: SFIP
ChlRC =Chltot 'Po 0
1 ÿ 'Po
1 ÿ 0
ChlRC =Chltot 'Po 0 SFIN 1 ÿ
ChlRC =Chltot 1 ÿ 'Po 1 ÿ 0 ChlRC 'Po 0 Chlantenna 1 ÿ 'Po 1 ÿ 0
The index 0 refers to the state at the onset of illumination. The PI de®ned here refers to photochemical events, an analogous expression denoted PIN can be de®ned for non-photochemical events where: PIN 1=PI:
2.3. Statistical analysis The distribution of all Chl a ¯uorescence data was tested for non-normality and heterocedasticity to ensure the assumptions were met for parametric statistics. The treatment eect was tested with a univariate analysis of variance (ANOVA), designed as a mixed model threestage nested using a type III sum of squares calculation. The highest level of the model was the treatment level with ®ve chambers nested in each treatment, and subgroups of provenances and trees, with subsubgroups of leaves within trees. Tests of signi®cance were made at the 95% con®dence level using a Tukey test. The SAS system for windows (Release 6.12, SAS Institute Inc., NC, USA) was used for all statistical analyses.
or in experimental terms:
RC FV 1 ÿ VJ ; PI VJ ABS F0 where Chltot is the total quantity of Chl a, and Chltot Chlantenna ChlRC . The ratio ChlRC/Chlantenna can be replaced by the ratio RC/ABS, where RC is the number of active PSII reaction centres, and ABS is the quantity of light absorbed by the antenna. RC/ABS, 'Po and 0 can be calculated according to the JIP-test using the experimentally collected parameters (Strasser and Strasser, 1995; Strasser et al., 1995). Maximum quantum yield of primary photochemistry: 'Po 1 ÿ F0 =FM TR0 =ABS
Density of reaction centres per chlorophyll: RC=ABS 'Po
VJ =M0 ;
where VJ is the relative variable ¯uorescence at 2 ms of the ¯uorescence rise, VJ
F2m s ÿ F0 =
FM ÿ F0 , and M0 is the slope at the origin of the relative variable ¯uorescence M0
F300 s ÿ F0 =
FM ÿ F0 . The value at 50 ms was used for F0. The probability of 0 that an electron is transported (ET) beyond Qÿ A is: 0 1 ÿ VJ ET0 =TR0 :
The ozone fumigation resulted in cumulative exposures (AOT40) of 0.01, 3.35, 7.06 and 19.7 ml lÿ1 h for the treatments I±IV, respectively. AOT40 was calculated for daylight hours (>50 Wmÿ2) using 1-h mean concentrations in excess of 40 nl lÿ1 ozone, in accordance with the UN/ECE de®nition (KaÈrenlampi and SkaÈrby, 1996) (Table 1). Daily levels of hourly mean ozone concentration in the 50% of ambient+50 nl lÿ1 were representative of those found at higher altitudes in the Swiss Alps. The repeated measures ANOVA calculated the eect of ozone on the PI between the 50% ambient+50 nl lÿ1 treatment and the other three treatments to be highly signi®cant over the entire season (P>0.0026) (Table 2). The PI values recorded from all plants remained indistinguishable until Day of Year 190, at which time the 50% ambient+30 nl lÿ1 ozone treatment PI values became increasingly reduced over time compared to the three other treatments (Fig. 1). The treatment PI means became signi®cantly dierent by the fourth measuring event (P>0.009) (Day of Year 226), and remained so for the rest of the season. By September, when the plants had received close to the complete seasons exposure, mean PI was reduced by an average of 70.5, 15.3, and 1% relative to the control (50% of ambient level). The development of visual damage was scored from 1, no visual symptom, to 5, large necrotic areas. The ®rst visual symptoms developed at about the same time as the mean PI value for trees in the 50% of ambient+ 30 nl lÿ1 treatment started to diverge from the other
A.J. Clark et al. / Environmental Pollution 109 (2000) 501±507
Table 2 The results of a repeated measures ANOVA, testing the between and within treatment eects of four ozone treatments on beech trees from two provenances using the primary photosynthetic performance index (PI) are showna df Between-treatment eects Treatment (error = *) Provenance (error = **) Treatmentprovenance** *Replication (Treatment) (error = **) Tree (Treatmentreplication rovenance) **Leaf Within-treatment effects Day of Year (DY) DYtreatment DYprovenance DYtreatmentprovenance DYreplication (Treatment) DYtree (Treatmentreplicationprovenance) Leaf (DY) error Huynh-Feldt "
3 1 3 15 15 150
21.586 0.106 1.041 2.855 2.53 0.082
0.0026 0.2577 0.0001 0.0001 0.0001 ±
7 21 7 21 105 105 1050
21.223 2.187 0.759 0.166 0.44 0.372 0.086
0.0001 0.0001 0.0001 0.0069 0.0001 0.0001 ± 1.0947
Replications refer to ®ve open-top chambers within each treatment. The degrees of freedom (df) and the mean square data are shown from which the F-test probabilities were calculated (P>F ). Asterisks indicate the speci®c error terms used at each level of the model.
Fig. 1. Performance index (PI) through the growing season (Day of Year). Four treatments are 50% ambient (!), 85% ambient (~), 100% ambient (*) and 50% of ambient+30 nl lÿ1 ozone (&). Error bars are calculated as 1 S.E.
treatments shown in Fig. 1. By Day of Year 210, some of the plants in the 50% of ambient+30 nl lÿ1 treatment were showing slight stippling on the leaf surfaces. The leaf symptoms became steadily more apparent throughout the season, until late August, when there were large necrotic areas. The mean PIN values demonstrated highly signi®cant correlation with the development of leaf damage (P>0.0001) (Fig. 2), with a linear regression r2 of 0.977. The visual symptoms can be viewed in terms of cumulative AOT40 exposure by
identifying the PIN values in Fig. 2 and reading from the corresponding values in Fig. 4 the AOT40 value. At the end of the season the mean weights of the seedlings were 1.91, 2.08, 1.81, and 1.30 g for the four treatments, respectively, as reported in full by Landolt et al. (2000). When compared to the control seedlings mean biomass per treatment was equivalent to a slight, but insigni®cant stimulation of 9% in the 85% of ambient ozone, an insigni®cant retardation of 5% in the ambient ozone and a highly signi®cant 32% reduction in the 50% ambient+30 nl lÿ1 ozone. Seedling weight was found to be highly correlated with mean PI values of the last 3 days measurements in September (Days 270, 273 and 279; Fig. 1) (Fig. 3). The signi®cant regression (r2=0.966) lacks biomass data between PI values of 0.3 and 0.8 because seedlings were harvested and biomass measured only at the end of the season. PI values between 0.3 and 0.8 are shown in Fig. 4 and would have been likely to ®t along the regression line were biomass data to have been collected throughout the season. PI demonstrates an ampli®ed sensitivity to the ozone exposure, and a degree of data noise, which is largely due to observed environmental dierences between the three measuring days (data not shown). The exposure response of beech primary photosynthesis performance is plotted in Fig. 4. PI and PIN are identical data sets and correspond with data previously illustrated in Figs. 1±3. The graph illustrates the eect of ozone independently of the large seasonal variation; all measuring dates were normalised by measuring treatment dierences relative to the control. The cumulative AOT40 ozone exposure the plants received at each successive measuring event versus the normalised
A.J. Clark et al. / Environmental Pollution 109 (2000) 501±507
Fig. 2. The inverse of the mean performance index (PIN) versus mean visual symptom score. PIN means were repeatedly measured during the season as visual symptoms developed. Error bars are calculated as 1 S.E.
Fig. 4. Performance index (PI) and the inverse of the mean performance index (PIN) versus cumulative AOT40 ozone exposure. Data points represent means repeatedly recorded during the 1997 season as the trees exposure to the seasons cumulative ozone increased. All ozone treatments are included. Error bars are calculated as 1 S.E.
Each point in Fig. 4 refers to the mean of 50 measured samples collected repeatedly throughout the growing season. For all treatments, a portion of the variance of the means was due to the signi®cant interaction that the tree provenances had with the Day of Year (Table 2). This is likely to be an adaptation to the local environmental conditions from which the groups of trees came. There was a highly signi®cant degree of variation between the individual trees (DYTree, Table 2). This probable genetic variation between individuals heavily outweighed the consistency recorded between leaves on the same tree. The stability of the error term at the leaf level resulted in a balanced data set, indicated by the Huynh-Feldt e. The broad variance between trees largely accounted for the broad variance apparent between OTC replicates because the number of tree replicates per OTC was only two. Fig. 3. End of season mean seedling biomass plotted versus performance index (PI), data have been normalised in respect to the 50% of ambient ozone treatment. Gridlines highlight control PI value of 1. Error bars are calculated as 1 S.E.
performance indices exposes the underlying response of primary photosynthesis (Fig. 4). PI and PIN can, with reference to the Figs. 1 and 2, be used to ®nd the visual symptoms induced and relative biomass changes correlated to a particular AOT40 exposure the trees received during the 1997 growing season (Fig. 4).
4. Discussion Exposure to ozone signi®cantly reduces the photosynthetic performance of beech over a single growing season, as measured by the PI index. This response demonstrated that the energy transduction process around PSII lost performance, con®rming previous research using Chl a ¯uorimetry (Mikkelsen, 1995; Mikkelsen and Heide-Jorgensen, 1996; Clark et al., 1998) and carbon dioxide gas exchange methods
A.J. Clark et al. / Environmental Pollution 109 (2000) 501±507
(Matyssek et al., 1991; Ruth and Weisel, 1993; Braun and FluÈckiger, 1995; KellomaÈki and Wang, 1997; Wieser, 1997; GuÈnthardt-Goerg et al., 1999). Higher values of PIN, the index for non-photochemical events, were highly correlated with the progressive development of visual symptoms of ozone damage re¯ected the increase in symptomatic energy dissipation through thermal losses. The divergence of PIN in the 50% ambient+30 nl lÿ1 ozone treatment from the other treatments occurred at the same time as the appearance of visual symptoms in July. The reduction of PI in the 50% ambient+30 nl lÿ1 ozone treatment relative to the other three treatments resulted in a reproducible correlation with the biomass of seedlings at the end of the season. Biomass increment and photosynthetic performance are likely to be constrained by similar physiological limitations and both can be considered measures of energy capture eciency. Cumulative stresses of ozone exposure were apparent in both relative biomass loss and the ¯uorescence measurements. The ¯uorescence emissions represent the active energetics occurring within seconds at the leaf tissue level while biomass is accumulated over months. More research will be needed to explain the reason for this correlation. The threshold value for the detection of the eect of ozone exposure on photosynthetic performance lay above the cumulative ambient values recorded in 1997. There were visible symptoms in the ambient treatment late in the season, but only in the 50%+30 nl lÿ1 treatment were they found to be signi®cant. In the attempt to determine an accurate method of ozone damage diagnosis that can be used regionally, the experimental results found signi®cant correlations between ozone exposure and photosynthetic performance index, PI. In the absence of a cost-eective ozone dose measurement, PI provides an alternative method for regional monitoring of tree health within the context of the currently employed AOT40. Once the PI, in combination with meteorological and ecological parameters, is more fully understood, detection and monitoring of forest areas at risk will be an achievable goal. Inclusion of the modifying in¯uence of the environment is central to modelling the damage due to ozone. The nature of leaf coupling to the atmosphere, stomatal control, water relations and ozone interception are all severely aected by the experimental situation (Wieser, 1997; Mans®eld, 1998), and have speci®c eects on the performance of photosynthesis (GruÈnhage and JaÈger, 1994). The high statistical signi®cance of the data show Chl a ¯uorescence measurements enable many samples to be screened in a short time. This method could be used in both Level I and Level II approaches to mapping critical levels of ozone as set out by KaÈrenlampi and SkaÈrby (1996).
Acknowledgement This work was supported by the Swiss National Science Foundation.
References Braun, S., FluÈckiger, W., 1995. Eects of ambient ozone on seedlings of Fagus sylvatica L. and Picea abies (L.) Karst. New Phytologist 129, 33±44. Clark, A.J., Landolt, W., Bucher, J.B., Strasser, R.J., 1998. The response of Fagus sylvatica to elevated CO2 and ozone probed by the JIP-test, based on the chlorophyll ¯uorescence rise: 0JIP. In: De Kok, L.J., Stulen, I. (Eds.), Responses of Plant Metabolism to Pollution. Backhuys Publishers, Netherlands, pp. 284±286. Clark, A.J., Landolt, W., Bucher, J.B., Strasser, R.J., 1999. Ozone exposure response of beech quanti®ed with a chlorophyll a ¯uorescence performance index. In: Fuhrer, J., Achermann, B. (Eds.), Critical Levels for Ozone Ð Level II. Environmental Documentation 115. Swiss Agency for the Environment, Forests and Landscape, Bern, Switzerland, pp. 177±180. Dann, M.S., Pell, E.J., 1989. Decline of activity and quantity of ribulose bisphosphate carboxylase/oxygenase and net photosynthesis in O3-treated potato foliage. Plant Physiology 91, 427±432. Farage, P.K., Long, S.P., Lechner, E.G., Baker, N.R., 1991. The sequence of changes within the photosynthetic apparatus of wheat following short term exposure to ozone. Plant Physiology 95, 529± 535. Furhrer, J., SkaÈrby, L., Ashmore, M.R., 1997. Critical levels for ozone eects on vegetation in Europe. Environmental Pollution 97, 91± 106. GruÈnhage, L., JaÈger, H.-J., 1994. In¯uence of the atmospheric conductivity on the ozone exposure of plants under ambient conditions: considerations for establishing ozone standards to protect vegetation. Environmental Pollution 85, 125±129. GuÈnthardt-Goerg, M.S., Matyssek, R., Scheidegger, C., Keller, T., 1993. Dierentiation and structural decline in the leaves and bark of birch (Betula pendula) under low ozone concentrations. Trees 7, 104±114. GuÈnthardt-Goerg, M.S., Maurer, S., Bolliger, J., Clark, A.J., Landolt, W., Bucher, J.B., 1999. Responses to near-ambient ozoneconcentrations of young trees (®ve deciduous tree species). Water, Air and Soil Pollution 116, 323±332. Heath, R.L., 1996. The modi®cation of photosynthetic capacity induced by ozone exposure. In: Baker, N.R. (Ed.), Photosynthesis and the Environment. Kluwer Academic, Netherlands, pp. 409±433. KaÈrenlampi, L., SkaÈrby, L. (Eds.), 1996. Critical Levels for Ozone in Europe: Testing and Finalising the Concepts. University of Kuopio, Department of Ecology and Environmental Science, Kuopio, Finland . KellomaÈki, S., Wang, K.-Y., 1997. Eects of elevated O3 and CO2 on chlorophyll ¯uorescence and gas exchange in Scots pine during the third growing season. Environmental Pollution 97, 17±27. Landolt, W., BuÈhlmann, U., Bleuler, P., Bucher, J.B., 2000. Ozone exposure±response relationships for biomass and root/shoot ratio of beech (Fagus sylvatica L.), ash (Fraxinus excelsior L.), Norway spruce (Picea abies [L.] Karst.) and Scots pine (Pinus sylvestris L.). Environmental Pollution 109, 473±478. Larcher, W., 1995. Photosynthesis as a tool for indicating temperature stress events. In: Schulze, E.-D., Caldwell, M.M. (Eds.), Ecophysiology of Photosynthesis. Springer-Verlag, Germany, pp. 261±277. Lichtenthaler, H.K., Rinderle, U., 1988. The role of chlorophyll ¯uorescence in the detection of stress conditions in plants. Critical Reviews in Analytical Chemistry 19, S29±S85.
A.J. Clark et al. / Environmental Pollution 109 (2000) 501±507 Mans®eld, T.A., 1998. Stomata and plant water relations: does air pollution create problems? Environmental Pollution 101, 1±11. Matyssek, R., GuÈnthardt-Goerg, M.S., Keller, T., Scheidegger, C., 1991. Impairment of gas exchange and structure in birch leaves (Betula pendula) caused by low ozone concentrations. Trees 5, 5±13. Mikkelsen, T.N., 1995. Physiological responses of Fagus sylvatica L. exposed to low levels of ozone in open-top chambers. Trees 9, 355± 361. Mikkelsen, T.N., Heide-Jorgensen, H.S., 1996. Acceleration of leaf senescence in Fagus sylvatica L. by low levels of tropospheric ozone demonstrated by leaf colour, chlorophyll ¯uorescence and chloroplast structure. Trees 10, 145±156. Reddy, G.N., Arteca, R.N., Dai, Y.R., Flores, H.E., Negram, F.B., Pell, E.J., 1993. Changes in ethylene and polyamines in relation to mRNA levels of the large and small subunits of ribulose bisphosphate carboxylase/oxygenase in ozone-stress potato foliage. Plant, Cell and Environment 16, 819±826. Ruth, B., Weisel, B., 1993. Investigations on the photosynthetic system of spruce aected by forest decline and ozone fumigation in closed chambers. Environmental Pollution 79, 31±35. Saxe, H., 1991. Photosynthesis and stomatal responses to polluted air, and the use of physiological and biochemical responses for early
detection and diagnostic tools. Annual Reviews in Botanical Research 18, 1±128. SkaÈrby, L., Ro-Poulsen, H., Wellburn, F.A.M., Sheppard, L.J., 1998. Impacts of ozone on forests: a European perspective. New Phytologist 139, 109±122. Strasser, B.J., Strasser, R.J., 1995. Measuring fast ¯uorescence transients to address environmental questions: the JIP test. In: Mathis, P. (Ed.), Photosynthesis: from Light to Biosphere, Vol. V. Kluwer Academic, Netherlands, pp. 977±980. Strasser, R.J., Srivastava, A., 1995. Govindjee, Polyphasic chlorophyll a ¯uorescence transient in plants and cyanobacteria. Photochemistry and Photobiology 61, 32±42. Strasser, R.J., Srivatava, A., Tsimilli-Michael, M., 1999. Screening the vitality and photosynthetic activity of plants by ¯uorescence transient. In: Behl, R.K., Punia, M.S., Lather, B.P.S. (Eds.), Crop Improvement for Food Security. SSARM, Hisar, India, pp. 72±115. Tsimilli-Michael, M., Pecheux, M., Strasser, R.J., 1998. Vitality and stress adaptation of the symbiont of coral reef and temperate foraminifers probed in hospite by the ¯uorescence kinetics OJIP. Archives de Science de GeneÁve 51, 205±240. Wieser, G., 1997. Ozone impact on photosynthetic capacity of mature and young Norway spruce (Picea abies (L.) Karst.): external versus exposure. Phyton 37, 297±302.