Journal of Experimental Botany, Vol. 52, No. 354, pp. 123±131, January 2001
Gas exchange by pods and subtending leaves and internal
recycling of CO2 by pods of chickpea (Cicer arietinum L.)
subjected to water deficits
Qifu Ma1, M.H. Behboudian2,3, Neil C. Turner1,2,4 and Jairo A. Palta1,2
1
Centre for Legumes in Mediterranean Agriculture, University of Western Australia, Nedlands, WA 6907,
Australia
2
CSIRO Plant Industry, Private Bag No. 5, Wembley, WA 6913, Australia
3
Institute of Natural Resources, Massey University, Palmerston North, New Zealand
Received 25 April 2000; Accepted 17 August 2000
Abstract
Terminal drought markedly reduces leaf photosynthesis of chickpea (Cicer arietinum L.) during seed
filling. A study was initiated to determine whether
photosynthesis and internal recycling of CO2 by the
pods can compensate for the low rate of photosynthesis in leaves under water deficits. The influence of
water deficits on the rates of photosynthesis and
transpiration of pods and subtending leaves in chickpea (cv. Sona) was investigated in two naturallylit, temperature-controlled glasshouses. At values
of photosynthetically active radiation (PAR) of
900 mmol m 2 s 1 and higher, the rate of net photosynthesis of subtending leaves of 10-d-old pods was
24 and 6 mmol m 2 s 1 in the well-watered (WW) and
water-stressed (WS) plants when the covered-leaf
water potential (C) was 0.6 and 1.4 MPa, respectively. Leaf photosynthesis further decreased to 4.5
and 0.5 mmol m 2 s 1 as C decreased to 2.3 and
3.3 MPa, respectively. At 900 ±1500 mmol m 2 s 1
PAR, the net photosynthetic rate of 10-d-old pods
was 0.9 ±1.0 mmol m 2 s 1 in the WW plants and was
0.1 to 0.8 mmol m 2 s 1 in the WS plants. The
photosynthetic rates of both pods and subtending
leaves decreased with age, but the rate of transpiration of the pods increased with age. The rates of
respiration and net photosynthesis inside the pods
were estimated by measuring the changes in the
internal concentration of CO2 of covered and
uncovered pods during the day. Both the WW and
WS pods had similar values of internal net photosynthesis, but the WS pods showed significantly
4
higher rates of respiration suggesting that the WS
pods had higher gross photosynthetic rates than
the WW pods, particularly in the late afternoon.
When 13CO2 was injected into the gas space inside
the pod, nearly 80% of the labelled carbon 24 h
after injection was observed in the pod wall in
both the WW and WS plants. After 144 h the proportion of 13C in the seed had increased from 19% to
32% in both treatments. The results suggest that
internal recycling of CO2 inside the pod may assist in
maintaining seed filling in water-stressed chickpea.
Key words: Pod photosynthesis, leaf photosynthesis, CO2
recycling, respiration, 13C labelling of pods.
Introduction
The rates of leaf photosynthesis in chickpea are markedly
reduced by water de®cits. Field measurements have shown
that they are below 5 mmol m 2 s 1 throughout seed ®lling
in a Mediterranean-type climate (Leport et al., 1998,
1999). However, seed ®lling continues under drought
conditions, albeit for a shorter duration (Davies et al.,
1999). Recent studies have suggested that the redistribution of dry matter from stems and leaves of chickpea
could provide up to 60% of the ®nal seed weight (Leport
et al., 1999), and that the pod wall could contribute
9±15% of the ®nal seed weight in chickpea grown under
terminal drought (Davies et al., 1999). More detailed
studies using 13C showed that less than 20% of the
seed carbon in chickpea came from pre-podding stored
assimilates when plants were water-stressed, proportions
To whom correspondence should be addressed at CSIRO. Fax: q61 8 9387 8991. E-mail: [email protected]
ß Society for Experimental Biology 2001
124
Ma et al.
signi®cantly less than those calculated from changes in
dry weight (Davies et al., 2000). Other studies using
14
CO2 suggested that the leaves were the primary source
of carbon for seed growth, particularly the leaf subtending the pod (Singh and Pandey, 1980). If the rate of
photosynthesis of this subtending leaf is very low, and this
needs to be veri®ed, and redistribution of carbon from
other plant parts is also low, the question arises as to the
source of carbon to maintain seed ®lling in water-stressed
chickpea.
Previous studies have shown that during the early
stages of seed development pod walls of chickpea can ®x
CO2 under well-irradiated conditions, but generate large
losses of CO2 through respiration in the dark (Sheoran
et al., 1987). In ®eld pea (Pisum sativum), it has been
shown that pod photosynthesis ®xed only a small amount
of atmospheric CO2 between 6 d and 30 d after ¯owering,
but under well-watered conditions re-®xed most of the
1500 to 15 000 ml l 1of CO2 respired during the day by
the seeds inside the pod cavity (Flinn et al., 1977). Similar
results were also reported in soybean (Glycine max)
(Sambo et al., 1977) and white lupin (Lupinus albus)
(Atkins and Flinn, 1978), indicating that re-®xing respired
CO2 inside the pod for use by the growing seed may be an
important mechanism for reducing losses of carbon.
Although photosynthesis by the external pod wall has
been suggested as contributing to seed growth in chickpea
(Sheoran et al., 1987), Leport et al. were unable to
measure any gas exchange by pods in the ®eld (Leport
et al., 1999). Moreover, the effects of water de®cits on the
gas exchange by pods and CO2 recycling inside the pod
remain unanswered. The aim of the present study was to
determine the in¯uence of water de®cits on the net CO2
exchange of both pods and subtending leaves at different
ages using more sensitive equipment than that of other
authors (Leport et al., 1999). Additionally, the role of the
pod wall and seed coat in internal CO2 ®xation and their
potential to reduce carbon losses during seed ®lling was
evaluated in well-watered and water-stressed chickpea.
Materials and methods
Four experiments were conducted. In Experiment 1, the gas
exchange of pods and subtending leaves at a range of light
intensities was measured with the increase in water de®cits and
age to establish whether the response of pods was similar to that
of leaves. In Experiment 2, diurnal changes in net CO2 exchange
of pods and internal levels of CO2 within the pods were
measured. Experiment 3 aimed at measuring the diurnal changes
in photosynthesis and respiration within the pod, and Experiment 4 examined the ®xation and distribution of labelled CO2
between the pod wall and seed over the ®rst 6 d after labelling.
Plant material, soil and fertilizer
The four experiments were conducted in two naturally-lit
glasshouses located at Floreat Park, Perth, Western Australia
(318579 S, 1158519 E) at various times of year in 1999. The
dayunight temperatures were maintained at 22u11 8C in one
glasshouse used for Experiments 1, 3 and 4, and at 22u15 8C
in the other glasshouse used for Experiment 2. Chickpeas
(Cicer arietinum L. cv. Sona) were grown in free-draining
polyvinylchloride pots (diameter 15 cm and depth 40 cm). The
seeds were sown in a commercial potting mix for germination,
and 1 week later one seedling was transplanted to each pot.
At transplanting, the roots were inoculated with commercial
group N Bradyrhizobium.
Experiments 1, 2 and 4 used the same soil and fertilizer. The
soil was a reddish brown, sandy clay loam (Calcic Haploxeralf,
pH 7.0 in 0.01 M CaCl2) collected from an area of undisturbed
native soil near Merredin, Western Australia. The soil was
sieved to remove any large aggregates and pieces of organic
matter. The top 10 cm of soil in the pots was a mixture of
Merredin soil and yellow sand at a ratio of 1:1 to prevent the soil
surface crusting, compared to a ratio of 9:1 of the same mixture
for the lower pro®le. Both the upper and lower soils were
mixed with fertilizers (g kg 1): 0.15 KNO3, 0.14 NH4NO3, 0.2
Ca(NO3)2, 0.15 triple superphosphate and 0.6 Micromax (g kg 1:
1.2 Fe, 0.25 Mn, 0.1 Zn, 0.05 Cu, 0.01 B, 0.005 Mo, and 1.5 S).
In Experiment 3, the soil was the surface 10 cm of a loamy sand
(Xeric Psamment, pH 4.6 in 0.01 M CaCl2) with 92% sand and
4% clay collected from a ®eld site near Wongan Hills, Western
Australia. The soil was mixed before transplanting with
fertilizers (g kg 1): 0.2 superphosphate (amended with Cu, Zn
and Mo) and 0.5 muriate of potash.
All pods were tagged when visible (3 mm) in order to select
pods of the same age at sampling. Plants began to set pods
about 8 weeks after sowing. At pod set, the subtending leaf was
about 15-d-old and ¯owering occurred 10 d prior to pod set.
Stomatal densities of leaves, and internal and external pod walls
were counted under a microscope (3 400) from acrylic surface
impressions from ®ve different pods or leaves. The surfaces were
sprayed with a thin coating of acrylic ®lm (artists ®xative) and
then transferred to a microscope slide on transparent tape
(Clemens and Jones, 1978).
Watering treatments
The soil in all four experiments was watered to ®eld capacity at
transplanting and then watered to saturation twice a week until
the two watering regimes were applied 70 d after sowing (DAS)
in Experiments 1, 2 and 4 and 75 DAS in Experiment 3. Water
was withheld from half of the plants, hereafter called waterstressed (WS), while the rest was kept watered as previously,
hereafter called well-watered (WW). In Experiments 1, 2 and 4,
the soil volume and soil type, and the use of only one plant per
pot, ensured a slow and steady development of water de®cits.
In these experiments pod and leaf measurements were made
between 11 d and 28 d after withholding water. In Experiment 3
water de®cits developed more rapidly in the sandy soil and
measurements were made 7 d after withholding water.
Measurements
Experiment 1 ± Light response of net photosynthesis and transpiration: The rates of net photosynthesis and transpiration
of pods and subtending leaves were measured using a CIRAS-1
portable photosynthesis system (PPS, Hitchin, UK) with
a standard leaf chamber which was large enough to enclose
a single pod. The system had a light unit which provided
levels of photosynthetically active radiation (PAR) from 10 to
1500 mmol m 2 s 1. Net photosynthetic measurements were
made at PARs of 10, 100, 300, 600, 900, 1200, and
Photosynthesis in water-stressed chickpea pods
2
1
1500 mmol m s . Both the pod and leaf photosynthetic rates
were calculated on a projected area basis. The leaf area : leaf
oven-dry weight ratio was measured on a representative set of
leaves in order to calculate the rate of photosynthesis on a leaf
dry weight basis. Pods of three different ages (10, 20 and 30 d
after pod set) were selected from both the WW and WS plants
with each measurement being replicated three times on different
plants. The rates of net photosynthesis and transpiration of
the pods and subtending leaves were measured 13, 18 and 26 d
after withholding water. On the same days the water potential
(C) of fully expanded leaves that had been covered with
aluminium foil on the late afternoon of the previous day was
measured near dawn using the pressure chamber technique,
following the precautions described previously (Turner, 1988).
Before measurement the proximal four lea¯ets were removed
and the leaf midrib was inserted into the pressure chamber.
Covered-leaf C is considered to be equivalent to the C in the
stem at the point of attachment to the leaf (Begg and Turner,
1970) and has been shown to be similar to the C of the pod
wall in chickpea (Shackel and Turner, 2000).
125
assumed that the rate of respiration in the light was the same
as that in the dark. The C of covered leaves was measured at
predawn, 09.00 h, 12.00 h, 15.00 h, and sunset by the pressure
chamber technique as in Experiment 1 with each measurement
being replicated ®ve times.
Experiment 4 ± 13CO2 feeding inside the pod: On day 11 after
withholding water the covered leaf water potential was
measured prior to dawn as in Experiment 1. At 09.00±10.00 h,
0.5 ml of 99.9% atom 13C-labelled CO2 was fed into the
cavity of 10±12-d-old pods of WW and WS plants. At feeding,
a hole was punctured in the basal end of a pod followed by
injection of 13CO2 in the proximal end using a 1 ml syringe,
and both holes were immediately sealed with a small drop of
glue that had no visible short-term or long-term effects on the
pod wall. The pods and subtending leaves were harvested at
24, 72 and 144 h after 13CO2 feeding and then quickly freezedried. After the pods had been separated into pod wall and
seed, the samples were weighed and ground to less than 1 mm
for analysis of 13C and C by mass spectrometry (VG
Micromass Sira 10) after Dumas combustion (Europa CuN
analyser).
Experiment 2 ± Pod net photosynthesis and internal CO2
concentration: Twenty-one days after withholding water, the
rates of net photosynthesis of 10±12-d-old pods were measured
at 06.00 h (predawn), 09.00 h, 12.00 h, 15.00 h, and 18.00 h
(sunset) on a clear sunny day using the CIRAS-1 portable
photosynthesis system. At the same time, the CO2 concentration
inside the pods of similar age was measured using a CI-301
(CID Inc., Moscow, ID, USA) CO2 analyser (Atkins and
Pate, 1977). The pods in the spaced chickpea plants were fully
exposed to light for most of the day. Before measurement, the
analyser was calibrated with four concentrations of CO2 ranging
from 0 to 1040 ml l 1. In 10±12-d-old pods, the internal gas
space is about 400 ml. During sampling, the needle of a 1 ml
syringe was inserted into the gas space in the proximal end
of a pod, and a 200 ml sample of gas was collected and
immediately injected into the analyser. Predawn covered leaf
C was measured using the pressure chamber technique as in
Experiment 1. All the measurements were replicated ®ve times
on different plants.
The pots were randomly arranged on benches in the glasshouse
and were moved weekly to ensure that positional effects were
removed. The rates of net photosynthesis and transpiration of
pods and their subtending leaves were analysed by a four-way
ANOVA using the water treatment (WW, WS), organ (pod,
leaf ), age (10, 20, 30 d) and light (10±1500 mmol m 2 s 1) as
main effects. The rates of respiration and net photosynthesis
inside the pod were analysed using a one-way ANOVA with
the water treatments as the main effect. GENSTAT 5.0 Release
3.2 was used for all ANOVA's.
Experiment 3 ± Net photosynthesis and respiration inside the
pod: The internal CO2 concentration of six 10-d-old pods per
treatment was measured from predawn (05.30 h) to sunset
(18.30 h) at intervals of approximately 2 h on a clear sunny
day, using a new set of pods each time and with the same
methods as described in Experiment 2. The decrease in CO2
concentration inside the pods between any two successive
measurements was used to estimate the rate of net photosynthesis
of the internal pod wall and seed coat for that period. At the
same time, a different group of six pods was covered with
aluminium foil for 2 h and the internal CO2 concentration
measured in order to estimate the rate of respiration by the
internal pod wall and seed coat for that period. In order to
establish the in¯uence of CO2 concentration on the rates of
net and gross photosynthesis inside the pod, another group of
six pods that had been covered for 2 h (resulting in high
internal CO2 concentrations) were uncovered and the CO2
concentration inside the pod sampled. Other pods were then
exposed to sunlight for 2 h before sampling. After the
measurement of CO2 concentration inside the pod, each pod
was harvested and its weight was measured after being dried
in a forced-draught oven at 70 8C for 72 h. The rates of
respiration and net photosynthesis inside the pod were
calculated on a dry weight basis and the sum of the two rates
was used to calculate the rate of gross photosynthesis. It was
Fully-expanded chickpea leaves had a stomatal density of
126"6 mm 2 compared to 31"3 mm 2 in the external
pod wall and no stomata in the internal pod wall.
Thirteen days after withholding water in Experiment 1, the covered-leaf C near dawn was 1.4 MPa in
the WS plants and 0.6 MPa in the WW plants. The
rates of net photosynthesis of 10-d-old pods were
0.9±1.0 mmol m 2 s 1 at PARs of 900±1500 mmol m 2 s 1
in the WW plants, and 0.1 to 0.3 mmol m 2 s 1 in the
WS plants (Fig. 1a). In contrast, the net photosynthetic
rate of the subtending leaves, which were 25-d-old at
the time of measurement, increased with increasing
PAR and reached 24 and 6 mmol m 2 s 1 at values of
PAR above 900 mmol m 2 s 1 in the WW and WS
plants, respectively (Fig. 1b). As the C fell to 2.3 and
3.3 MPa with time in the WS plants, the pods respired
and their subtending leaves ®xed CO2 at the rates of 4.5
and 0.5 mmol m 2 s 1, respectively.
In the WW plants the rate of pod transpiration was
0.55 mmol m 2 s 1 at PARs of 1200 to 1500 mmol m 2 s 1
and decreased to half of that when the covered-leaf
Statistical analyses
Results
Pod and leaf gas exchange
126
Ma et al.
Fig. 1. Response of rates of net photosynthesis (a, b) and transpiration (c, d) of 10-d-old pods (a, c) and their subtending leaves (b, d) to
photosynthetically active radiation (PAR) at four different levels of covered-leaf water potential (C) measured near dawn. The vertical lines give the
values of l.s.d. (P ˆ 0.05).
C fell to 1.4 MPa (Fig. 1c). In contrast, the rate
of leaf transpiration in the WW chickpeas reached
a maximum value of 3.4 mmol m 2 s 1 at a PAR of
900 mmol m 2 s 1, but decreased to 1.0 mmol m 2 s 1
when the C dropped to 1.4 MPa (Fig. 1d). Both
pods and subtending leaves had lower transpiration
rates with the further development of water de®cits.
The rates of net photosynthesis and transpiration of
the pods and subtending leaves were also affected by age.
While 10- and 20-d-old pods had positive rates of
photosynthesis at high values of PAR, 30-d-old pods
always respired irrespective of PAR (Fig. 2a). Subtending
leaves of 10- and 20-d-old pods, which were 25-d-old and
35-d-old at the time of measurement, had similar light
response curves for CO2 ®xation, but the 45-d-old leaves
showed light saturation at a PAR of 600 mmol m 2 s 1
and the maximum rate of net photosynthesis was reduced
to half that in the younger leaves (Fig. 2b). The rate of
leaf transpiration was also signi®cantly lower (one-third)
in the old leaves compared to that in the young leaves
(Fig. 2d). In contrast, the rate of pod transpiration was
the highest in the 30-d-old pods and lowest in the 10-d-old
pods (Fig. 2c).
Diurnal changes in pod net photosynthesis
and internal CO2 levels
Twenty-one days after withholding water in Experiment 2, predawn leaf C was 0.7 MPa in the WW
plants and 1.7 MPa in the WS plants. At 06.00 h
(predawn) 10±12-d-old pods were respiring, but from
09.00±15.00 h the WW pods ®xed CO2 at rates between
0.3 and 1.0 mmol m 2 s 1 (Fig. 3a), within the range of
the standard errors in Experiment 1. In the WS plants,
the rate of net photosynthesis of the pods was lower
than that of the WW pods near midday. At 18.00 h
(sunset), the pods were again respiring at rates greater
than those at predawn, consistent with results in ®eld pea
(Atkins and Pate, 1977) and presumably because the
glasshouse was warmer at 18.00 h (day temperature)
than at 06.00 h (night temperature). The CO2 concentration inside the pods (Fig. 3b) was almost the inverse
Photosynthesis in water-stressed chickpea pods
127
Fig. 2. Response of rates of net photosynthesis (a, b) and transpiration (c, d) of pods (a, c) and subtending leaves (b, d) of well-watered chickpea to
photosynthetically active radiation (PAR) at three different ages. Pods were set when the subtending leaves were 15-d-old. The vertical lines give the
values of l.s.d. (P ˆ 0.05).
of the net photosynthesis of the pod measured by external
gas exchange. At predawn the CO2 concentration inside
the WW and WS pods was about 4500 ml l 1, but
decreased to near zero at 09.00 h. The internal CO2
concentration then increased and was higher in the WS
than the WW pods. At sunset, the CO2 concentration
inside the WS pods was 11 000 ml l 1, but only 7000 ml l 1
in the WW pods. Similar results were obtained in
a further experiment.
Diurnal changes in respiration
and photosynthesis inside the pod
Seven days after withholding water in Experiment 3, the
predawn leaf C was 0.6 and 1.8 MPa in covered
leaves and decreased to 0.8 and 2.3 MPa during the
day in the WW and WS plants, respectively (Fig. 4a). The
internal rate of respiration of 10-d-old pods was higher in
the WS plants than the pods of the WW plants throughout the day, particularly in the afternoon (Fig. 4b).
The rate of net photosynthesis inside the pods was
0.10±0.14 mmol g 1 s 1 from predawn to 08.00 h with the
WS plants being higher than the WW plants (Fig. 4c).
During the rest of the day, the rate of net photosynthesis inside the pods was either negative or low
(0.01±0.02 mmol g 1 s 1) and was not signi®cantly different between the WW and WS pods (Fig. 4c). As
a consequence of the higher rates of respiration in the
WS pods, the pods had higher calculated rates of gross
photosynthesis inside the pods than the WW pods
throughout the day (Fig. 4d). To determine the effect of
high CO2 concentrations on the internal pod photosynthesis, the concentration of CO2 inside the pod was
elevated by covering the pods for 2 h prior to the
measurements of photosynthesis. The pods with elevated
initial levels of CO2 had higher net and gross photosynthetic rates inside the pods in both the WW and WS
plants (data not shown). The data from both the WW and
WS pods with normal and elevated initial levels of CO2
showed that there was a positive linear relationship
(r2 ˆ 0.81) between the estimated rate of gross photosynthesis and the initial CO2 concentration inside the
128
Ma et al.
differences between the treatments in
redistribution inside the pods.
13
C ®xation or
Discussion
Fig. 3. Diurnal changes of pod net photosynthetic rate (a) and CO2
concentration inside the pods (b) of the well-watered (m) and waterstressed (k) chickpea. The measurements were taken from 10±12-d-old
pods. Bars give "SE of the mean (n ˆ 5).
pods (Fig. 5). The rate of gross photosynthesis was
always higher at the same internal CO2 concentration
in the WS than WW pods, possibly due to the greater
seed size (0.263"0.003 g seed 1 in the WS and 0.174"
0.001 g seed 1 in the WW chickpeas) and faster rate of
growth in the WS seeds than the WW seeds.
Allocation of
13
CO2 inside the pod
Eleven days after withholding water, the predawn coveredleaf C was 0.5 MPa for the WW plants and 1.4 MPa
for the WS plants. Twenty-four hours after 0.5 ml of
13
CO2 was fed into the cavity of 10±12-d-old chickpea
pods in Experiment 4, the excess 13C was similar at
55.6"4.8 and 58.3"6.1 mg in the WW and WS pods,
respectively, but no measurable 13C was observed in the
subtending leaves. At this time the proportion of 13C in
the pod wall was 78% and 80% in the WW and WS pods,
while the proportion in the seed was 22% and 20% in the
WW and WS pods, respectively (Fig. 6). Two days later
a greater proportion of the 13C had moved to the seeds,
and this trend continued so that by 6 d after feeding only
65% and 64% of the 13C was in the pod wall, 31% and
34% in the seed, and 4% and 2% in the leaves in the WW
and WS pods, respectively. There were no signi®cant
The present study showed that leaf net CO2 exchange
decreased with both a reduction in water potential and
leaf age and that the leaves were less responsive to light
when water-stressed. These results are consistent with
those observed previously, showing that when the midday leaf C in rainfed chickpea fell below 3 MPa leaf
photosynthesis decreased to a tenth of its maximum
(Leport et al., 1998, 1999). The decrease in leaf photosynthesis with water stress appeared to be largely mediated
through stomatal closure as shown from the transpiration
data. In contrast to a previous study with less sensitive
gas exchange equipment which showed no net CO2
exchange by pods in the ®eld (Leport et al., 1999), this
study found that the net CO2 exchange of young, wellirradiated and well-watered pods was positive at values of
PAR above 200 mmol m 2 s 1. However, the rate of pod
respiration increased with age and the response of the
pods to light decreased with water stress so that older
pods, especially when water stressed, had negative rates of
net CO2 exchange. Stomata were observed on the external
wall of the pods, although these were fewer in number
than those on the leaves. The transpiration data showed
that the stomata on the pod wall were responsive to light
and water de®cits in a similar manner to those on the
leaves. However, in contrast to leaf stomata, the stomata
on the external pod wall became more leaky with age
leading to higher rates of transpiration and poorer water
use ef®ciencies in old compared to young pods.
While the young and well-watered pods ®xed CO2
from the atmosphere outside the pod, the rates of gas
exchange were at best only 5% of those in subtending
leaves. This suggests that the gas exchange by the pod
is only a minor contributor to seed ®lling in chickpea.
Saxena and Sheldrake also came to the same conclusion
from experiments in which chickpea pods were exposed to
full sunlight during seed ®lling (Saxena and Sheldrake,
1980). According to Flinn et al., the fruit's respiratory
output at night in the well-watered ®eld pea was always
in excess of its daytime gain of CO2 from the atmosphere
so that overall it was in negative balance for carbon
throughout its life (Flinn et al., 1977). Low carbon ®xation by pods has also been reported in common bean
(Crookston et al., 1974), and rapeseed (Singh, 1995).
Sheoran et al. found that covering the pod with
aluminium foil from day 3 after ¯owering reduced seed
dry weight by 20% and concluded that the pod CO2
exchange contributed 20% to the seed yield (Sheoran
et al., 1987). This conclusion assumes that all the CO2
uptake by the pod is utilized for seed growth, whereas in
Photosynthesis in water-stressed chickpea pods
129
Fig. 4. Diurnal changes in covered leaf water potential (a), rate of respiration (b), net photosynthesis (c), and gross photosynthesis (d) inside intact
pods of chickpeas that were either well watered or water stressed. The pods were 10±12-d-old. Bars give one standard error of the mean (n ˆ 6) except in
(a) where bars give "SE of the mean (n ˆ 6) for values greater than the size of the symbol.
Fig. 5. Relationship between the rate of gross photosynthesis and the
initial CO2 concentration inside 10±12-d-old chickpea pods. Pods
were either uncovered (m, k) or covered (m, n) prior to the
measurement and were sampled from both well-watered (m, m) and
water-stressed (k, n) plants.
early pod development most of the dry matter is used for
the growth of pod wall (Davies et al., 1999). In addition,
when the pod is covered, not only ®xation of atmospheric
CO2 is inhibited, but also any CO2 recycling inside the
pod would be reduced.
The measurements in this study indicate that the gas
exchange with the atmosphere by the pods is unlikely to
compensate for the stress-induced reduction in current
assimilate for seed growth. As a previous study has shown
that the carbon accumulated during vegetative growth
by the pod wall, leaves, stems, and roots and redistributed
to the seed contributed only 20% of the carbon in the
seed (Davies et al., 2000), there is still a considerable
proportion of carbon needed for seed growth that is
unaccounted for from these sources. Previous studies
have suggested that the inner epidermis of the pod wall
in well-watered pulses and the embryo of rapeseed is
effective in re®xing CO2 respired by the seeds (Flinn
and Pate, 1970; Flinn et al., 1977; Sheoran et al., 1987;
King et al., 1998), thereby reducing losses of carbon
by respiration. In common bean, the total CO2-®xing
capacity through recycling within the pod was estimated
130
Ma et al.
Fig. 6. Allocation of excess 13C between pod wall, seed and subtending
leaf after the pod cavity was fed with 0.5 ml of 99.9% atom 13C-labelled
CO2 in (a) well-watered and (b) water-stressed chickpeas. Bars give one
SE of the mean (n ˆ 6).
to be 26% of that of the subtending leaf (Crookston et al.,
1974), and in ®eld pea the inner pod wall was estimated to
be capable of re®xing 66% of the CO2 released by the
seeds into the gas space in the light (Atkins et al., 1977).
However, studies have not previously been undertaken in
pods of water-stressed pulses. In Experiment 2, the CO2
concentration inside the pods changed diurnally, showing
high values at 06.00 h (predawn) and 18.00 h (sunset), but
lower values during the day particularly in the morning.
The decrease of CO2 concentration inside the pods during
the day clearly resulted from CO2 ®xation within the pod
and not by the leakage to the external atmosphere since
the pod showed positive net CO2 gain from the atmosphere at times when the CO2 concentration inside the
pod was decreasing. The decrease in CO2 between 06.00 h
and 09.00 h in Experiment 2 indicates that the pods ®xed
0.9 mg carbon per pod during this 3 h period. The ®xation
at this rate occurred in both the WW and WS plants.
Indeed, the rate of CO2 recycling (gross photosynthesis)
inside the pod of 0.7±1.0 mmol g 1 s 1 near midday in the
WW and WS plants, respectively, compares favourably
with the rate of net photosynthesis of 0.8 mmol g 1 s 1
in the leaves of the WW chickpea when measured at the
same time of day and expressed on the same dry weight
basis.
The similarity in carbon ®xation inside the pod of
WW and WS plants was also seen after 13CO2 was injected
into the cavity of young pods, suggesting that the recycling
of CO2 in the pod was unaffected when chickpea plants
were subjected to water-stress. Indeed, the detailed diurnal
studies of CO2 concentrations inside the pod in Experiment
3 showed that gross photosynthesis inside the pods was at
least as high in the WS than as the WW plants, possibly as
a result of the higher internal rates of respiration arising
from the faster rates of initial seed growth in the WS
pods (Davies et al., 1999). Covering the pods to generate
high initial CO2 concentrations inside the pods resulted in
high CO2 ®xation rates in both the WW and WS plants.
The linear relationship between gross photosynthesis and
initial CO2 concentration inside the pods (Fig. 5) suggests
that the internal CO2 recycling was very ef®cient and the
low rate of respiration of the external pod wall in the
dark (Fig. 1) suggests that there is a high resistance to
CO2 diffusion within the pod wall itself. The recycling of
CO2 inside the pod under water-stressed conditions presumably reduced carbon losses to the atmosphere and
increased the utilization of carbon, allowing continued seed
growth even though the plants were subjected to terminal
drought. However, the re-®xation of CO2 respired by
the seeds depends on a supply of carbon from elsewhere,
such as the redistribution from the leaves and stems (Davies
et al., 2000) and from the low but positive rates of photosynthesis in the leaves of water-stressed chickpea that
were observed in this study and in the ®eld (Leport et al.,
1998, 1999).
The ef®ciency of CO2 recycling within the pods, even
when plants were water-stressed, may be related to
the morphology and water relations of the pod and seed.
The high density of chloroplasts in the inner epidermis, the
thin cuticle and the dome-shaped outer contours of
the epidermal cells of the internal pod wall in ®eld pea
have been interpreted as specializations for photoassimilation of CO2 from the pod cavity (Atkins et al., 1977).
While the inner epidermis of the chickpea pod wall does
not have any stomata on its surface and the data suggest
that there is a high resistance to diffusion of CO2 in the
pod wall, the 13C study showed that the pod wall initially
®xed about 80% of the carbon and over time this carbon
was redistributed to the seed. The redistribution of 10%
of the 13C from the pod wall to the seed over the ®rst
6 d supports the previous observation that pod walls
contribute to seed ®lling during rapid seed growth in
chickpea, particularly in water-stressed chickpea (Davies
et al., 1999).
Photosynthesis in water-stressed chickpea pods
In light of the fact that the pod wall has water potentials similar to those in the leaves of WS plants and much
lower than those in the seed (Shackel and Turner, 2000),
the observation that nearly 80% of the carbon is ®xed by
the pod wall in WS as well as WW chickpeas suggests that
low water potentials do not inhibit carbon ®xation in the
inner pod wall. Alternatively, considerable variation may
exist in water potentials across the pod wall in the waterstressed chickpea that are not detected by the pressure
chamber technique and not observed previously (Shackel
and Turner, 2000). These authors only measured the
turgor of the cells about 400 mm from the outer epidermis
of the pod. The apoplastic space would need to be near
full saturation for the seed coat turgor to be maintained
(Shackel and Turner, 1998). Therefore, the cells in the
pod wall adjacent to the apoplastic space where ®xation
likely took place (Atkins et al., 1977) may have been
considerably higher in water potential than the average of
the pod wall measured by the pressure chamber technique
or that measured by the pressure probe near the outer
epidermis. The variation in water status across the pod
wall and the site of ®xation of CO2 inside the pod wall
are worthy of further investigation.
This study has shown that while water stress suppressed the net CO2 exchange with the atmosphere by
both leaves and pods during seed ®lling, high rates of CO2
recycling inside the pods may help to compensate for the
low rates of photosynthesis in the leaves and provide an
important source of carbon for seed growth, additional to
the previously-reported carbon redistribution from pod
walls, leaves and stems, in water-stressed chickpea
(Leport et al., 1999; Davies et al., 1999, 2000).
Acknowledgements
We are grateful to Dr Don White for the use of the CIRAS-1
portable photosynthesis system. We also thank Jens Berger for
statistical assistance, Christiane Ludwig, Elaine Smith, Mike
Barr, and Rebecca Kenney for their technical assistance and
Drs KHM Siddique, RJ French, RT Furbank, Z Rengel, and
SL Davies for comments on the manuscript. MHB thanks
CSIRO for the award of a McMaster Fellowship and the Massey
University for the provision of leave.
References
Atkins CA, Kuo J, Pate JS. 1977. Photosynthetic pod wall of
pea (Pisum sativum L.), distribution of carbon dioxide-®xing
enzymes in relation to pod structure. Plant Physiology 60,
779±786.
Atkins CA, Flinn AM. 1978. Carbon dioxide ®xation in the
carbon economy of developing seeds of Lupinus albus (L.).
Plant Physiology 62, 486±490.
Atkins CA, Pate JS. 1977. An IRGA technique to measure CO2
content of small volumes of gas from the internal atmospheres
of plant organs. Photosynthetica 11, 214±216.
131
Begg JE, Turner NC. 1970. Water potential gradients in ®eld
tobacco. Plant Physiology 46, 343±346.
Clemens J, Jones PG. 1978. Modi®cation of drought resistance
by water stress conditioning in Acacia and Eucalyptus. Journal
of Experimental Botany 29, 895±904.
Crookston RK, O'Toole J, Ozbun JL. 1974. Characterization of
the bean pod as a photosynthetic organ. Crop Science 14,
708±712.
Davies SL, Turner NC, Siddique KHM, Plummer JA, Leport L.
1999. Seed growth of desi and kabuli chickpea (Cicer
arietinum L.) in a short-season Mediterranean-type environment. Australian Journal of Experimental Agriculture 39,
181±188.
Davies SL, Turner NC, Palta JA, Siddique KHM, Plummer JA.
2000. Remobilization of carbon and nitrogen supports seed
®lling in chickpea subjected to water de®cit. Australian
Journal of Agricultural Research 51, 855±866.
Flinn AM, Atkins CA, Pate JS. 1977. Signi®cance of photosynthetic and respiratory exchanges in the carbon economy of
the developing pea fruit. Plant Physiology 60, 412±418.
Flinn AM, Pate JS. 1970. A quantitative study of carbon
transfer from pod and subtending leaf to the ripening seeds of
the ®eld pea (Pisum arvense L.). Journal of Experimental
Botany 21, 71±82.
King SP, Badger MR, Furbank RT. 1998. CO2 ®xation
characteristics of developing canola seeds and silique wall.
Australian Journal of Plant Physiology 25, 377±386.
Leport L, Turner NC, French RJ, Barr MD, Duda R, Davies SL,
Tennant D, Siddique KHM. 1999. Physiological responses of
chickpea genotypes to terminal drought in a Mediterraneantype environment. European Journal of Agronomy 11,
279±291.
Leport L, Turner NC, French RJ, Barr MD, Tennant D,
Thomson BD, Siddique KHM. 1998. Water relations, gas
exchange and growth of cool-season grain legumes in
a Mediterranean-type environment. European Journal of
Agronomy 9, 295±303.
Sambo EY, Moorby J, Milthorpe FL. 1977. Photosynthesis and
respiration of developing soybean pods. Australian Journal
of Plant Physiology 4, 713±721.
Saxena NP, Sheldrake AR. 1980. Effects of pod exposure
on yield of chickpeas (Cicer arietinum). Field Crops Research
3, 189±191.
Shackel KA, Turner NC. 1998. Seed coat cell turgor responds
rapidly to air humidity in chickpea and faba bean. Journal
of Experimental Botany 49, 1413±1419.
Shackel KA, Turner NC. 2000. Seed coat cell turgor is
unresponsive to irrigation despite changes in plant and pod
water potential in chickpea. Journal of Experimental Botany
51, 895±900.
Sheoran IS, Singal HR, Singh R. 1987. Photosynthetic
characteristics of chickpea (Cicer arietinum L.) pod wall
during seed development. Indian Journal of Experimental
Biology 25, 843±847.
Singh BK, Pandey RK. 1980. Production and distribution of
assimilate in chickpea (Cicer arietinum L.). Australian Journal
of Plant Physiology 7, 727±735.
Singh R. 1995. Regulation of CO2 assimilation in pod walls of
Brassica campestris L. Haryana Agricultural University,
Hisar, India.
Turner NC. 1988. Measurement of plant water status by the
pressure chamber technique. Irrigation Science 9, 289±308.