ORIGINAL ARTICLE
Photosynthetic CO2 uptake in seedlings of two tropical tree species
exposed to oscillating elevated concentrations of CO2
Received: 24 January 2003 / Accepted: 7 July 2003 / Published online: 6 August 2003

 Springer-Verlag 2003
Abstract Do short-term ?uctuations in CO2 concentra-
tions at elevated CO2 levels a?ect net CO2 uptake rates
of plants? When exposed to 600 ll CO2 l
)1, net CO2
uptake rates in shoots or leaves of seedlings of two
tropical C3 tree species, teak (Tectona grandis L. f.) and
barrigon [Pseudobombax septenatum (Jacq.) Dug.], in-
creased by 28 and 52% respectively. In the presence of
oscillations with half-cycles of 20 s, amplitude of ca.
170 ll CO2 l
)1 and mean of 600 ll CO2 l
)1, the stimu-
lation in net CO2 uptake by the two species was reduced
to 19 and 36%, respectively, i.e. the CO2 stimulation in
photosynthesis associated with a change in exposure
from 370 to 600 ll CO2 l
)1 was reduced by a third in
both species. Similar reductions in CO2-stimulated net
CO2 uptake were observed in T. grandis exposed to 40-s
oscillations. Rates of CO2 e?ux in the dark by whole
shoots of T. grandis decreased by 4.8% upon exposure
of plants grown at 370 ll CO2 l
)1 to 600 ll CO2 l
)1. The
potential implications of the observations on CO2
oscillations and dark respiration are discussed in the
context of free-air CO2 enrichment (FACE) systems in
which short-term ?uctuations of CO2 concentration are
a common feature.
Keywords CO2 oscillations ? Elevated CO2
concentrations ? Free-air CO2 enrichment (FACE)
systems ? Photosynthesis ? Tropical trees
Abbreviations FACE: free-air CO2 enrichment ? IRGA:
infra-red gas analyser ? Rubisco: ribulose 1,5-bis-
phosphate carboxylase/oxygenase
Introduction
Free-air carbon dioxide enrichment (FACE) facilities
are presently considered the best method for manipu-
lating atmospheric CO2 concentrations around plants
growing under otherwise natural ?eld conditions
(Hendrey et al. 1999; McLeod and Long 1999). Such
systems have become an integral tool for studying, in the
context of global climate change, the e?ects of increasing
CO2 concentrations on the growth and development of
uncontained plants in situ (Miglietta et al. 2001).
All FACE systems impose two CO2 treatments ? an
increase in the average CO2 concentration and a ?uc-
tuating, often oscillating, CO2 treatment. The amplitude
and frequency of the variations in CO2 concentration
common in FACE systems are usually much greater
than would ever be experienced under natural condi-
tions, even near tropical forest ?oors (Holtum and
Winter 2001). Changes in CO2 partial pressures of
200?300 ll CO2 l
)1 over periods of 5?20 s are not
uncommon in FACE systems but ?uctuations of 300 ll
CO2 l
)1 for periods of 30 s or longer are rare (Evans and
Hendrey 1992). Such estimates of ?uctuations may be
underestimates as long sampling lines may dampen the
signals and some ?uctuations may be faster than moni-
tor response times.
The CO2 concentrations ?uctuate because the CO2
injection mechanisms overshoot or undershoot as they
continually adjust to counteract variations in wind speed
and direction. For example, CO2 concentrations based
on over one million 1-s measurements (each an integral
of 3 s) over a 2-year period at the University of
Arizona?s Maricopa Agricultural Center FACE facility
were more than 110 ll l)1 higher or lower than the set
target of 550 ll CO2 l
)1 for 9.3% of the time (Nagy et al.
1994). For 23.9% of the time they di?ered by between
Planta (2003) 218: 152?158
DOI 10.1007/s00425-003-1089-1
Joseph A. M. Holtum ? Klaus Winter
Owing to an unfortunate misunderstanding, the uncorrected
version of this paper was published.
J. A. M. Holtum ? K. Winter (&)
Smithsonian Tropical Research Institute,
P.O. Box 2072, Balboa, Ancon,
Republic of Panama
E-mail: winterk@tivoli.si.edu
Fax: +507-2128148
Present address: J. A. M. Holtum
Tropical Plant Sciences,
James Cook University, Townsville,
Queensland 4811, Australia
?55 and ?110 ll l)1. When averaged over 1-min
intervals, a common measure of FACE performance, the
CO2 concentration was within ?55 ll l
)1 of the set
point for 95% of the time and between ?55 and
?110 ll l)1 for 6.7% of the time. The CO2 ?uctuations
induced in Brookhaven-type FACE facilities that inject
diluted CO2 (e.g. Nagy et al. 1994; Hendrey et al. 1999;
Jordan et al. 1999) and in comparably sized facilities
that inject pure CO2 (Miglietta et al. 2001; Okada et al.
2001; Pepin and Ko?rner 2002) are broadly similar al-
though FACE systems enclosing trees and natural
communities tend to exhibit greater ?uctuations than
those enclosing crops. Open-top systems can exhibit
comparable ?uctuations (Cardon et al. 1995; Winter
et al. 2000).
Clearly, FACE systems do not mimic atmospheric
CO2 conditions over time-scales of a few minutes or less
(with the exception of natural CO2 vents; Koch 1993;
Miglietta et al. 1993). Although there is extensive liter-
ature on the e?ects of constant high CO2 concentrations
on plant growth and development, there have been few
studies that compare the e?ects on net CO2 uptake and
plant performance of rapidly oscillating versus constant
CO2 concentrations. Do short-term oscillations in CO2
concentration a?ect photosynthetic CO2 exchange in the
shorter term and plant growth in the longer term?
Although it is commonly expressed that such short-term
variations are unimportant in situ (Hendrey et al. 1997,
1999), particularly in tree species for which the responses
of stomata are believed to be slower than in crop plants
(Saxe et al. 1998), short-term CO2 oscillations have been
reported to perturb photosynthesis in leaves of the C3
species Gossypium hirsutum L. (Evans and Hendrey
1992), Triticum aestivum L. (Hendrey et al. 1997) and
Phaseolus vulgaris L. (Cardon et al. 1994, 1995) and in
the C4 species Zea mays L. (Cardon et al. 1994, 1995).
However, in none of the above-mentioned examples was
CO2 uptake studied during oscillations of less than
1 min that are characteristic of FACE experiments.
Gossypium leaf tissue exposed to 1-min oscillations of
between 360 and 1,090 ll CO2 l
)1 (mean of 700 ll l)1)
exhibited a mean rate of uptake of 14CO2 that did not
di?er from that of leaf tissue which had been exposed to
a constant concentration of 700 ll l)1 (Evans and
Hendrey 1992). However, oscillations of 2 min and
longer were associated with an increase in net CO2 gain,
reaching 27% when the oscillation was extended to
10 min. It was speculated that the mechanism respon-
sible for the increase was related to postulated changes
from ribulose 1,5-bisphosphate carboxylase/oxygenase
(Rubisco)-limited to inorganic phosphate- and triose
phosphate-limited photosynthesis. Furthermore, it was
suggested by extrapolation that oscillations of less than
1 min duration would have little e?ect on the rate of
long-term carbon gain.
Photosynthetic CO2 uptake in wheat was inferred
from measurements of instantaneous photosystem II
?uorescence (Ft) during oscillations with an amplitude of
225 ll CO2 l
)1 around a mean of 575 ll CO2 l
)1 and
half-cycles between 0.1 and 64 s (Hendrey et al. 1997).
Oscillations in chlorophyll ?uorescence were observed
for half-cycles greater than 2 s and reductions in electron
transport rate (J) were observed for half-cycles of 30 s
and greater. It was concluded that at least 180 s were
required before Ft signals achieved a new steady state,
and that a substantial decrease in CO2 uptake would
occur only if the duration of a CO2 oscillation was
greater than 1 min, or if the oscillation was not sym-
metric around the mean.
In Z. mays and P. vulgaris subjected to CO2 oscilla-
tions of 100?160 ll CO2 l
)1 for between 2 and 20 min,
stomatal conductance shifted away from the steady-state
level observed under the median CO2 concentration of
333?340 ll CO2 l
)1 (Cardon et al. 1994, 1995). The
extent and direction of the shifts, which depended upon
the species and the oscillation frequency, were related to
species-speci?c di?erences in the kinetics of stomatal
movement and photosynthetic characteristics. The non-
steady-state conditions changed short-term water-use
e?ciencies in both species although photosynthetic rates
remained fairly constant.
In order to dispel uncertainty on the e?ects of short-
term ?uctuations in CO2 concentrations on carbon gain
we have tested whether the responses of net CO2
exchange by seedlings or leaves of two tropical tree
species, teak (Tectona grandis L. f.) and Pseudobombax
septenatum (Jacq.) Dug., to an increase in CO2 concen-
tration from ca. 370 to 600 ll CO2 l
)1 are a?ected by
symmetric oscillations around 600 ll CO2 l
)1, with half-
cycles of considerably less than 1 min.
Exposure to enhanced and ?uctuating CO2 are not
the only treatments imposed by FACE systems. A
number of FACE systems impose a third CO2 treatment:
the CO2 injectors are turned o? during the dark (Pepin
and Ko?rner 2002). Apart from reducing the use and thus
the cost of CO2, switching o? the CO2 supply avoids the
technical problem of controlling and maintaining con-
stant and relatively uniform CO2 concentrations when
wind speeds are low, and reduces blower-induced can-
opy temperature increases (Pinter et al. 2000). There is
uncertainty as to whether plant performance and
development is a?ected by increased concentrations of
CO2 in the dark, a period when photosynthesis is not
taking place and ambient concentrations of CO2 tend to
be higher. Although dark respiration by C3 and C4
grasses, C3 herbaceous species and C3 trees has been
reported to be inhibited under enhanced CO2 concen-
trations (e.g. Drake et al. 1999), there are many reports
of little or no e?ect of enhanced CO2 concentrations on
dark respiration (e.g. Amthor et al. 2001; Hamilton et al.
2001; Tjoelker et al. 1999, 2001). Recently it has been
suggested that some reports on the e?ects of high CO2
on dark respiration may be artefacts caused by the
leakage of CO2 from plant gas-exchange chambers
through gaskets or through contiguous pores which
connect regions of plant mesophyll that transcend the
boundaries of the chambers (Jahnke 2001; Jahnke and
Krewitt 2002; Pons and Welschen 2002).
153
In order to quantify dark CO2 e?ux in T. grandis and
to circumvent problems associated with the leakage of
respiratory CO2 through leaves or across gaskets we
determined the e?ects of an increase inCO2 concentration
from 370 to 600 ll CO2 l
)1 on dark respiration by whole
intact shoots of teak seedlings that were fully enclosed in a
gas-exchange chamber.
Materials and methods
Plant material and growth conditions
Seeds of Tectona grandis L. f. (Verbenaceae) and Pseudobombax
septenatum (Jacq.) Dug. (Bombacaceae) were collected locally and
germinated in potting soil in a screenhouse on the roof of the Tupper
Building, Smithsonian Tropical Research Institute, Panama City,
Republic of Panama. After 2?3 weeks, seedlings were transplanted
into half-strength Johnson?s solution (Winter 1973) and grownunder
a 12 h light, 26 C/12 h dark, 23 C cycle in an environmental
growth chamber (GCT-8; GEC, Chagrin Falls OH, USA) equipped
with eight ?uorescent light tubes (Sylvania 115 W F48T12/CW/
VHO).
Gas exchange system
Net CO2 exchange was measured for the shoots of whole plants in a
through-?ow gas exchange system (Walz, E?eltrich, Germany).
Oscillating CO2 concentrations were generated by mixing two air
streams, one containing CO2 and the other containing CO2-free air.
The CO2-containing air stream was generated by mixing pure CO2
and CO2-free air in a custom-made mixing unit (Walz GMA-3). The
CO2-free streamwas generatedbypassing air through soda-lime.The
dew-points of the twoair streamswere set by electronically controlled
cold-traps (Walz KF-24/6BM and KF-18/2) before passage through
two mass-?ow controlled pumps (Walz LD-5R and LD-10R). Air
streams with oscillating CO2 concentrations were generated by
alternating the supply from each pump at appropriate intervals using
a timer-controlled solenoid gas switch (Walz TG 101A and Walz
GUS-8). Air was pumped through a Plexiglas cuvette with a volume
of 1.21 l (11 cm ? 11 cm ? 10 cm). Mixing of the atmosphere inside
the cuvette was facilitated by a 4-cm-diameter CPU cooler fan (12 V,
0.08 A). The airstream leaving the cuvette was dehumidi?ed in a
cold-trap at 2 C (Walz KF-18/2) and the CO2 concentration
determined by an infra-red gas analyser (IRGA; LI-6252; LI-COR,
Lincoln, NE, USA) previously calibrated using CO2 gas standard
(Scott Speciality Gases, Plumsteadville PA, USA) and a set of three
gas-mixing pumps (Wo?stho?, Bochum, Germany). Gas ?ow rates
were 2.200, 2.128, 1.100 and 1.032 l min)1 for experiments at con-
stant CO2 in the light, for 20-s oscillations, for 40-s oscillations and
for experiments at constant CO2 during the dark, respectively. Flow
rates were veri?ed using a water-volume displacement method and a
digital soap-bubble ?ow meter (model 650; Humonics Inc, Rancho
Cordova CA, USA).
In an experiment designed to test the dilution of oscillation
signals in the airstream between the plant chamber and the IRGAwe
compared maximum and minimum CO2 concentrations emanating
from the complete gas-exchange system with the signals emanating
from the system when the post-chamber pre-IRGA cold-trap had
been removed and the IRGA was directly connected to the outlet of
the gas-exchange chamber. The dilution of the extremes of the
oscillations averaged 7 ll l)1.
Measurements of net CO2 exchange
Intact seedlings of ca. 6 cm height, growing in 150-ml pots
containing half-strength Johnson?s solution (Winter 1973) were
inserted into a gas-exchange cuvette located in the temperature-
controlled growth chamber in which the seedlings had been
maintained. For T. grandis, the entire shoot was sealed in the
cuvette (total leaf area of 44?77 cm2), whereas for P. septenatum
one leaf was enclosed (area of 29?33 cm2). Plant material in the
cuvette was kept under a regime of 12 h light, 29 C/12 h dark,
25 C. The dew-point of the air entering the gas-exchange cuvette
was 18 C. The light intensity at the uppermost leaf inside the
cuvette was 280 lmol m)2 s)1 for the experiments with T. grandis
and 410 lmol m)2 s)1 for P. septenatum.
Plant material was incubated at 370 ll CO2 l
)1 in the gas-ex-
change cuvette overnight. Experiments were initiated about 2 h
following the onset of the light period. After determining net CO2
exchange rate at a constant 370 ll CO2 l
)1, the CO2 concentration
was increased to 600 ll CO2 l
)1 and net CO2 exchange was re-
corded following attainment of steady-state photosynthesis. For
the experiments with oscillating CO2 concentrations, gas exchange
was recorded for 10 min in the presence of the plant tissue and then
for 10 min in the absence of the plant tissue. Estimations of net
CO2 exchange did not alter when the sequence of collecting sample
and control data was reversed. The output from the gas analyser
was sampled electronically at 1-s intervals. CO2 uptake by the tis-
sue was calculated from the di?erence in the integrated CO2 con-
centrations and expressed as a mean rate per second on a leaf-area
basis.
To obtain CO2?response curves of net CO2 exchange in the
light, CO2 concentrations were increased in three steps from 370 to
850 ll CO2 l
)1, decreased in six steps to 30 ll CO2 l
)1 and then
increased in ?ve steps to 600 ll CO2 l
)1. Each CO2 concentration
was maintained until a steady-state rate of photosynthesis was
attained.
Dark respiration rates were determined during the normal dark
period. Measurements were taken at 370 ll CO2 l
)1, at 600 ll
CO2 l
)1 and subsequently at 370 ll CO2 l
)1.
Results
The rate of net CO2 uptake in the light by T. grandis was
CO2-dependent (Fig. 1). When exposed to a constant
concentration of 600 ll CO2 l
)1, the rate of net CO2
uptake was 28?3% (mean ? SE) greater than at a
constant 370 ll l)1 (P ? 0.01, paired t-test; columns 3
and 4 in Table 1). This CO2-dependent increase at
Fig. 1 CO2?response curve of a whole shoot of a Tectona grandis
seedling exposed to constant concentrations of CO2 (open circles)
or to oscillations in CO2 concentration with half-cycles of 20 s
(closed circle). Representative of four experiments on three plants
154
600 ll CO2 l
)1 was reduced to 19?3% (SE; P ? 0.01,
paired t-test) when the tissue was exposed to symmetric
oscillations with a mean of 600 ll CO2 l
)1, a half-cycle
of 20 s and an amplitude of ca. 170 ll CO2 l
)1 (Fig. 2,
Table 1). Similarly, in the subset of plants exposed to 40-
s oscillations, the 30?3% (SE; P ? 0.01, paired t-test)
increase of net CO2 uptake was reduced to 20?2% (SE;
P ? 0.01, paired t-test). That is, in every experiment
performed with T. grandis under oscillating CO2
conditions of less than 1 min, net CO2 uptake diminished.
The reduction of the stimulation of photosynthetic CO2
uptake associated with the increase from 370 to 600 ll
CO2 l
)1 was 36?5% (SE) in the presence of oscillations
with a 20-s half-cycle and 34?6% (SE) in the presence of
oscillations with a 40-s half-cycle (Table 1).
Similar observations were made for photosynthetic
CO2 uptake by P. septenatum (Table 2, Fig. 3). The
stimulation in net CO2 exchange in response to an
increase in the CO2 concentration from a constant
370 ll CO2 l
)1 to a constant 600 ll CO2 l
)1 was 52?2%
(SE; P ? 0.01, paired t-test). This increase was reduced
to 36?2% (SE; P ? 0.01, paired t-test) when the tissue
was exposed to 20-s oscillations, i.e. reduction of the
stimulation of CO2 uptake was 31?3% (SE) in the
presence of 20-s oscillations (Table 2).
For shoots of T. grandis the rates of respiratory net
CO2 loss during the dark were examined at 25 C at a
constant 370 and a constant 600 ll CO2 l
)1 (Table 3).
The rate of net CO2 production in the presence of 600 ll
CO2 l
)1 averaged 4.8?1.3% (SE) less than the average
of the rates at 370 ll CO2 l
)1. The di?erences were sig-
ni?cant at a level of P ? 0.01 (paired t-test).
Discussion
The potential for short-term ?uctuations in CO2 con-
centration, typical of FACE systems, to alter photo-
synthetic carbon gain from that observed under constant
CO2 concentrations has been commented upon a num-
ber of times (Evans and Hendrey 1992; Cardon et al.
1994, 1995; Nagy et al. 1994; Hendrey et al. 1997;
McLeod and Long 1999; Pepin and Ko?rner 2002). Rates
of net CO2 exchange under rapidly ?uctuating CO2
Table 1 Net CO2 uptake by shoots of Tectona grandis seedlings at
constant and oscillating CO2 concentrations. Percentage reduction
in CO2-stimulated CO2 uptake was calculated as [(D)E)/
(D)C)?100], where the capital letters indicate values in the col-
umns from left to right. The means of rates for 20-s and 40-s
oscillations did not di?er from each other (pairedt-test), but dif-
fered from rates at constant 370 and 600 ll CO2 l
)1 (P ? 0.01,
pairedt-test). The rates at constant 370 and 600 ll CO2 l
)1 di?ered
(P ? 0.01, paired t-test)
Plant No. Expt. No. Net CO2 uptake (lmol m
)2 s)1) Reduction in
CO2-stimulated
rate under oscil-
lating CO2(%)
Constant Constant Oscillating
370 ll CO2 l
)1 600 ll CO2 l
)1 600 ll CO2 l
)1
20 s 40 s 20 s 40 s
1 1 4.98 6.90 6.13 ? 40.1 ?
2 ? 6.46 5.77 ? ? ?
3 ? 6.26 5.81 ? ? ?
4 5.01 7.35 6.75 ? 25.6 ?
2 1 6.56 7.96 7.37 ? 42.1 ?
2 7.26 7.77 7.64 ? 25.5 ?
3 7.19 8.68 7.59 ? 73.2 ?
4 7.61 9.32 8.37 8.36 55.6 56.1
5 7.35 9.07 8.64 8.66 25.0 23.8
3 1 4.99 6.58 6.20 6.08 23.9 31.4
2 4.74 6.67 6.58 5.96 5.2 36.8
3 6.05 7.71 7.04 7.44 40.4 16.3
4 5.90 7.78 7.13 7.08 34.6 37.2
Fig. 2 CO2 concentrations experienced by the shoot of the
T. grandis seedling illustrated in Fig. 1 during 10 complete
oscillations each with a half-cycle of 20 s. CO2 concentration was
sampled every 1 s. The mean CO2 concentration during the
experiment depicted was 598.9?0.2 ll CO2 l
)1, the mean of the
maxima was 766.0?0.2 ll CO2 l
)1 and the mean of the minima was
433.5?0.1 ll CO2 l
)1 (values ? SE). Similar regular kinetics were
observed during experiments with oscillations of 40 s half-cycle
155
concentrations have not been measured in real time
because of the technical di?culty of accurately esti-
mating the di?erences between the rapidly changing CO2
concentrations in the reference and sample airstreams.
We circumvented this problem by separating in time the
measurements of the reference and sample airstreams
and integrated the rapidly changing CO2 concentrations
in both airstreams over a number of oscillations
(10 min). Mean rates of net CO2 exchange could thus be
calculated, and treatments at oscillating and constant
CO2 concentrations could be compared.
In bothT. grandis andP. septenatum, rapid oscillations
of CO2 at frequencies and amplitudes commonly experi-
enced by vegetation inside FACE systems consistently
reduced by about a third the increase in net carbon gain
associated with an increase in CO2 concentration from
370 to 600 ll CO2 l
)1 (Tables 1, 2). Oscillations in atmo-
spheric CO2 should only in?uence photosynthetic rate in
C3 plants if the concentration of dissolved CO2 at the site
of the Rubisco is altered and if the activity of the
carboxylase is limited by CO2 at some point during the
oscillation. Change in the concentration of CO2 at the site
of Rubisco should re?ect the amplitude and frequency of
oscillation, and the rate at which carbon di?uses from the
atmosphere to the chloroplast. The oscillations will be
dampened as CO2 traverses the boundary layer, passes
through the stomate into the sub-stomatal cavity,
dissolves in the cell milieu and di?uses to the chloroplast.
Clearly, the photosynthetic CO2-assimilating appa-
ratus in T. grandis and P. septenatum can respond to
extremely rapid changes in external CO2 concentration.
Analogous rapid responses have been reported in wheat
for measurements of ?uorescence under non-photore-
spiratory conditions (Hendrey et al. 1997). Chlorophyll
?uorescence yield (Ft) in wheat leaves responded to half-
cycles as short as 2 s when exposed to oscillations of
amplitude 225 ll CO2 l
)1 around a mean of 575 ll
CO2 l
)1, and electron transport through photosystem II
(J) was reduced by about 10% when exposed to 30-s
half-cycles and 20% when exposed to half-cycles of 60 s
or greater oscillating around a mean of 650 ll CO2 l
)1
with an amplitude of 215 ll CO2 l
)1.
A model has been proposed to explain the decrease in
photosynthetic net carbon gain in the presence of
oscillating CO2 concentrations (see Fig. 1 in Hendrey
et al. 1997). The model assumes that the concentrations
of CO2 within the oscillating range fall within the
partially saturated portion of the photosynthetic CO2?
response curve (see Fig. 1), and that during oscillations
Table 2 Net CO2 uptake by leaves of Pseudobombax septenatum
seedlings at constant and oscillating concentrations of CO2.
Percentage reduction in CO2-stimulated CO2 uptake was calculated
as [(D)E)/(D)C)?100], where the capital letters indicate values in
table columns from left to right. The means of the rates at constant
370, constant 600 ll CO2 l
)1 and for 20-s oscillations di?ered from
each other (P ? 0.01, paired t-test)
Plant No. Expt. No. Net CO2 uptake (lmol m
)2 s)1) Reduction in CO2-stimulated
rate under oscillating CO2(%)
Constant Constant Oscillating
370 ll CO2 l
)1 600 ll CO2 l
)1 600 ll CO2 l
)1 20 s
1 1 8.44 12.76 10.89 43.3
2 7.21 10.70 9.77 26.6
2 1 6.11 9.53 8.47 31.0
3 1 7.00 10.86 9.99 22.5
4 1 5.66 8.36 7.35 37.4
2 7.03 10.93 10.02 23.3
Table 3 Net CO2 production by shoots of ?ve seedlings of Tectona
grandis during the dark. Plants were sequentially exposed to 370,
600 and 370 ll CO2 l
)1. The rate of net CO2 production in the
presence of 600 ll CO2 l
)1 averaged 4.8?1.3% (SE) less than the
average of the rates at 370 ll CO2 l
)1. The di?erences were
signi?cant at a level of P ? 0.01 (paired t-test)
Plant No. Net CO2 production (lmol m
)2 s)1)
370 ll CO2 l
)1 600 ll CO2 l
)1 370 ll CO2 l
)1
1 0.695 0.685 0.720
2 0.805 0.785 0.850
3 0.740 0.715 0.740
4 0.740 0.690 0.730
5 0.825 0.780 0.840
Fig. 3 CO2?response curve of a leaf of a Pseudobombax septenatum
seedling exposed to constant concentrations of CO2 (open circles)
or to oscillations in CO2 concentration with half-cycles of 20 s
(closed circle). The mean CO2 concentration during the oscillation
experiment depicted was 599?2 ll CO2 l
)1, the mean of the
maxima was 775?2 ll CO2 l
)1 and the mean of the minima was
440?3 ll CO2 l
)1 (values ? SE)
156
the leaf tissue is exposed to the maximum and minimum
oscillatory concentrations of CO2 for a duration su?-
cient to permit steady-state photosynthesis to occur, i.e.
the oscillations are rectangular in shape. Under such
conditions the mean of the two extreme steady-state
rates of photosynthesis will lie below the curve. Our
observations with T. grandis and P. septenatum are
consistent with the model in that net carbon gain under
short-term oscillations fell below the curve (Figs. 1, 3).
However, the situation is more complex because in our
experiments, which were designed to emulate FACE
conditions, the external CO2 concentration changed
continuously, never reaching a steady state.
We have not examined the e?ects of rapid oscillations
on stomatal aperture, a response that can indirectly af-
fect photosynthetic carbon gain. Cardon et al. (1994,
1995) demonstrated in Zea mays and Phaseolus vulgaris
that the average stomatal conductance during 3- to
20-min oscillations with medians of 333?340 ll CO2 l
)1
and amplitudes of 100?160 ll CO2 l
)1 could be driven
far from the steady-state condition observed at the
median CO2 concentration. Both the extent and the
direction of the departure from the steady state was
dependent upon species-speci?c asymmetries in stomatal
opening and closing kinetics as well as the frequency and
amplitude of oscillations in CO2.
A small but consistent reduction of 4.8% in respira-
tory carbon loss was observed at constant 600 ll CO2 l
)1
in comparison to that observed at a constant 370 ll CO2
l)1. It is unlikely that the reduction in carbon loss is the
result of leakage of CO2 from tissues reported by Jahnke
and Krewitt (2002) and Pons and Welschen (2002) as the
entire shoot of each T. grandis plant was enclosed in the
gas-exchange chamber and the stem was tightly sealed
with the non-porous synthetic rubber sealant Terostat
VII (Henkel-Teroson, Heidelberg, Germany), rather
than a semi-porous gasket. Similarly, one cannot ascribe
the small di?erences in respiratory loss to changes in the
water vapour content of the airstream, which was
dehumidi?ed in an electronically controlled water
vapour trap at 2 C prior to IRGA analysis.
The decrease in the rate of respiratory carbon loss
from the shoots from T. grandis, which was measured
during the normal dark period of the plants at 25 C,
was about 2-fold that reported for 12 C3 and C4 grass-
land species (Tjoelker et al. 2001), 35?70% of that
observed for sweetgum, Liquidambar styraci?ua
(Hamilton et al. 2001), and about one-third of that re-
ported by Amthor (1997) who analysed the data for 36
species in 45 studies. The decrease in respiratory CO2
loss in T. grandis is small enough to be accounted for by
direct e?ects of CO2 on mitochondrial enzymes (Drake
et al. 1999). Respiratory CO2 loss at night may be
reduced by phosphoenolpyruvate carboxylase (PEPC) as
is the case for weak crassulacean acid metabolism
(CAM) plants (Holtum and Winter 1999) but in non-
CAM plants, doubling the ambient CO2 concentration is
unlikely to a?ect the rates of net CO2 loss in the dark via
PEPC (Melzer and O?Leary 1987; Amthor 1997).
The calculated increase in 24-h carbon gain associ-
ated with the change from 370 to 600 ll CO2 l
)1 was
29.8% when 600 ll CO2 l
)1 was only provided during
the daylight hours, and 30.4% when 600 ll CO2 l
)1 was
also provided at night. In the context of FACE experi-
ments, this reduction in dark respiration observed in
seedlings of the C3 plant T. grandis represents a tri?ing
increase in net carbon gain. However, bearing in mind
the variety of values published for the e?ects of
increasing CO2 concentration on respiratory dark loss
(see Amthor 1997; Drake et al. 1997; Curtis and Wang
1998) there is clearly a need for further studies on whole
intact plants rather than leaf segments or detached
leaves.
In conclusion, we have demonstrated that short-term
oscillations in CO2 concentration matter. However, it is
unclear whether the responses are species-speci?c,
whether plant CO2 exchange acclimates to oscillating
CO2 in the long-term, whether the reduction in net
carbon gain persists and, if so, whether the reduction
translates into reduced growth. Moreover, in our
experiments the oscillations were regular in periodicity
and uniform in amplitude and shape, although this is
not the case in FACE systems. Even so, our observa-
tions raise the possibility that FACE systems may
underestimate the potential fertilising e?ects of above-
ambient CO2 concentrations on plants.
Acknowledgements This research was funded by the Andrew W.
Mellon Foundation and the Smithsonian Tropical Research Insti-
tute. We thank Milton Garcia for skilled technical assistance.
J.A.M.H. acknowledges support from Dr. R.G. Dunn and a
Queensland-Smithsonian Fellowship.
References
Amthor JS (1997) Plant respiratory responses to elevated CO2
partial pressure. In: Allen LH, Kirkham MB, Olszyk DM,
Whitman CE (eds) Advances in carbon dioxide e?ects research.
American Society of Agronomy Special Publication (Proceed-
ings of 1993 ASA Symposium, Cincinatti, Ohio). ASA, CSSA
and SSSA, Madison, WI, pp 35?77
Amthor JS, Koch GW, Willms JR, Layzell DB (2001) Leaf O2
uptake in the dark is independent on coincident CO2 partial
pressure. J Exp Bot 52:2235?2238
Cardon ZG, Berry JA, Woodrow IE (1994) Dependence of the
extent and direction of average stomatal response in Zea mays
L. and Phaseolus vulgaris L. on the frequency of ?uctuations in
environmental stimuli. Plant Physiol 105:1007?1013
Cardon ZG, Berry JA, Woodrow IE (1995) Fluctuating [CO2]
drives species-speci?c changes in water use e?ciency. J Bioge-
ogr 22:203?208
Curtis PS, Wang X (1998) A meta-analysis of elevated CO2 e?ects
on woody plant mass, form, and physiology. Oecologia
113:299?313
Drake BG, Muehe MS, Peresta G, Gonza`lez-Meler MA, Matamala
R (1997) Acclimation of photosynthesis, respiration and eco-
system carbon ?ux of a wetland on Chesapeake Bay, Maryland
to elevated CO2 concentration. Plant Soil 187:111?118
Drake BG, Azcon-Bieto J, Berry J, Bunce J, Dijkstra P, Farrar J,
Gi?ord RM, Gonza`lez-Meler MA, Koch G, Lambers H (1999)
Does elevated atmospheric CO2 concentration inhibit mito-
chondrial respiration in green plants? Plant Cell Environ
22:649?657
157
Evans LS, Hendrey GR (1992) Responses of cotton foliage to
short-term ?uctuations in CO2 partial pressures. Crit Rev Plant
Sci 11:203?212
Hamilton JG, Thomas RB, DeLucia EH (2001) Direct and indirect
e?ects of elevated CO2 on leaf respiration in a forest ecosystem.
Plant Cell Environ 24:975?982
Hendrey GR, Long SP, McKee IF, Baker NR (1997) Can photo-
synthesis respond to short-term ?uctuations in atmospheric
carbon dioxide? Photosynth Res 51:179?184
Hendrey GR, Ellsworth DS, Lewin KF, Nagy J (1999) A free-air
enrichment system for exposing tall forest vegetation to ele-
vated atmospheric CO2. Global Change Biol 5:293?309
Holtum JAM, Winter K (1999) Degrees of crassulacean acid
metabolism in tropical epiphytic and lithophytic ferns. Aust J
Plant Physiol 26:749?757
Holtum JAM, Winter K (2001) Are plants growing close to the
?oors of tropical forests exposed to markedly elevated con-
centrations of carbon dioxide? Aust J Bot 49:629?636
Jahnke S (2001) Atmospheric CO2 concentration does not directly
a?ect leaf respiration in bean or poplar. Plant Cell Environ
24:1139?1151
Jahnke S, Krewitt M (2002) Atmospheric CO2 concentration may
directly a?ect leaf respiration measurement in tobacco, but not
respiration itself. Plant Cell Environ 25:641?651
Jordan DN, Zitzer SF, Hendrey GR, Lewin KF, Nagy J, Nowak
RS, Smith SD, Coleman JS, Seemann JR (1999) Biotic, abiotic
and performance aspects of the Nevada Desert free-air CO2
enrichment (FACE) facility. Global Change Biol 5:659?668
Koch G (1993) The use of natural situations of CO2 enrichment in
studies of vegetation responses to increasing atmospheric CO2.
In: Schulze ED, Mooney HA (eds) Design and execution of
experiments on CO2 enrichment. Commission of the European
Communities, Brussels, Belgium
McLeod AR, Long SP (1999) Free-air carbon dioxide enrichment
(FACE) in global change research: a review. Adv Ecol Res
28:1?56
Melzer E, O?Leary MH (1987) Anapleurotic CO2 ?xation by
phosphoenolpyruvate carboxylase in C3 plants. Plant Physiol
84:58?60
Miglietta F, Raschi A, Battarini I, Resti R, Selvi F (1993) Natural
CO2 springs in Italy: a resource for examining long-term
response of vegetation to rising atmospheric CO2 concentra-
tions. Plant Cell Environ 16:873?878
Miglietta F, Peressotti A, Vaccari FP, Zaldei A, deAngeles P,
Scarascia-Mugnozza G (2001) Free-air CO2 enrichment
(FACE) of a poplar plantation: the POPFACE fumigation
system. New Phytol 150:465?476
Nagy J, Lewin KF, Hendrey GR, Hassinger E, LaMorte R (1994)
FACE facility CO2 concentration control and CO2 use in 1990
and 1991. Agric For Meteorol 70:31?48
Okada M, Lie?ering M, Nakamura H, Yoshimoto M, Kim HY,
Kobayashi K (2001) Free-air CO2 enrichment (FACE) using
pure CO2 injection: system description. New Phytol 150:251?
260
Pepin S, Ko?rner C (2002) Web-FACE: a new canopy free-air CO2
enrichment system for tall trees in mature forests. Oecologia
133:1?9
Pinter PJ, Kimball BA, Wall GW, LaMorte RL, Hunsaker DJ,
Adamsen FJ, Frumau KFA, Vugts HF, Hendrey GR, Lewin
KF, Nagy J, Johnson HB, Wechsunge F, Leavitt SW,
Thompson TL, Matthias AD, Brooks TJ (2000) Free-air CO2
enrichment (FACE): blower e?ects on wheat canopy microcli-
mate and plant development. Agric For Meteorol 103:319?333
Pons TL, Welschen RAM (2002) Overestimation of respiration
rates in commercially available clamp-on leaf chambers. Com-
plications with measurement of net photosynthesis. Plant Cell
Environ 25:1367?1372
Saxe H, Ellsworth DS, Heath J (1998) Tree and forest functioning
in an enriched CO2 atmosphere. New Phytol 139:395?436
Tjoelker MG, Reich PB, Oleksyn J (1999) Changes in leaf nitrogen
and carbohydrates underlie temperature and CO2 acclimation
of dark respiration in ?ve boreal tree species. Plant Cell Environ
22:767?778
Tjoelker MG, Oleksyn J, Lee TD, Reich PB (2001) Direct inhibi-
tion of leaf dark respiration by elevated CO2 is minor in 12
grassland species. New Phytol 150:419?424
Winter K (1973) CO2-Fixierungsreaktionen bei der Salzp?anze
Mesembryanthemum crystallinum unter variierten Au?enbe-
dingungen. Planta 114:75?85
Winter K, Garcia M, Lovelock CE, Gottsberger R, Popp M (2000)
Responses of model communities of two tropical tree species to
elevated atmospheric CO2: growth on unfertilised soil. Flora
195:289?302
158