1. Stem Water Potential is a Sensitive Indicator of Grapevine Water Status
XAVIER CHONEÂ {, CORNELIS VAN LEEUWEN*{{, DENIS DUBOURDIEU{
and JEAN PIERRE GAUDILLEÁ RE}
{Faculte d'Oenologie de Bordeaux, Universite Bordeaux 2 Victor SeÂgalen, 351 Crs de la LibeÂration, 33405 Talence
Cedex, France, {ENITA de Bordeaux, 1 Crs du GeÂneÂral de Gaulle, 33175 Gradignan Cedex, France and }INRA
Agronomie, Station de la Grande Ferrade, 71 av Edouard Bourleau, 33140 Villenave d'Ornon, France
Received: 31 August 2000 Returned for revision: 3 November 2000 Accepted: 13 December 2000 Published electronically: 23 February 2001
Dawn leaf water potential (dawnC), leaf water potential (leafC) and stem water potential (stemC) were measured on
mature leaves to determine non-irrigated vine water status in vineyards during the growing season. StemC was the
most discriminating indicator for both moderate and severe water de®cits. The di€erence between stemC and leafC
(DC) provided an indirect measurement of mean leaf transpiration which varied with soil moisture conditions and
vapour pressure de®cit in the atmosphere. The use of stemC as an indicator for grapevine management in both non-
irrigated and irrigated vineyards is discussed. # 2001 Annals of Botany Company
Key words: Water status, vine, root zone, stem water potential, leaf water potential, transpiration ¯ow, hydraulic
conductivity.
INTRODUCTION
The current soil-plant-atmosphere-continuum (SPAC)
model is mainly based on the theory that water must be
under tension to be transported through the plant's xylem.
This model explains the occurrence of water de®cits along
the pathway from the roots to the leaves. Plant water
transport follows schematically in four steps: soil to root;
root to shoot xylem; shoot to leaf through the petiole; and
leaf to atmosphere through stomata. Mean plant water
status depends on water potential in soil layers close to the
root system, canopy size and evaporative demand. Internal
plant water de®cits occur to ®t xylem sap ¯ow to leaf
transpiration in relation to soil water availability.
Vine water status is an important factor in grape quality.
High tannin and anthocyanin content in red grape berries is
related to moderate vine water de®cits (Matthews et al.,
1990; Van Leeuwen et al., 1994).
To determine the in¯uence of environmental and cultural
conditions on vine water status, a sensitive physiological
indicator that integrates both soil and climatic conditions is
required. The pressure chamber (Scholander et al., 1965) is
a reliable method for determining the water status of ®eld-
grown vines. Pressure chamber measurements can provide
values of dawn leaf water potential (dawnC), daily leaf
water potential (leafC) and stem water potential (stemC).
Predawn leaf water potential measures plant water status
at zero plant water ¯ux and provides information on the
root zone soil water potential because predawn plant water
status is considered to be in equilibrium with soil water
status. Daily leafC measured on a single leaf re¯ects a
combination of many factors: local leaf water demand
[vapour pressure de®cit (VPD), leaf intercepted radiation];
soil water availability; internal plant hydraulic conductivity
and stomatal regulation. StemC is measured on a non-
transpiring leaf (Begg and Turner, 1970). Daily stemC is the
result of whole plant transpiration, and soil and root/soil
hydraulic conductivity. StemC indicates the capacity of
grapevine to conduct water from the soil to the atmosphere.
StemC has been successfully applied as a water de®cit
indicator on peach and plum orchards (Garnier and Berger,
1985; MacCutchan and Shackel, 1992). Liu et al. (1978) and
Greenspan et al. (1996) measured stemC and water
potential gradients in vines. In peach trees, the di€erence
between stemC and leafC (DC ˆ stemC À leafC)
measured simultaneously on the same plant was shown to
be a indicator of instantaneous shoot transpiration.
Assuming constant hydraulic conductivity in the petioles
throughout the growing season, DC varies with soil water
availability in the root zone (Garnier and Berger, 1985).
Here we reconsider stem water potential as a noticeable
indicator of plant water de®cit. The relationship between
leaf transpiration and DC is established for grapevines.
MATERIALS AND METHODS
Experimental location and plant material
This study was conducted in 1998 in a Californian vineyard
(Dominus Estate, Yountville, Napa Valley, USA) and in a
French vineyard (Haut Brion, Pessac, Bordeaux, France).
LeafC and stemC were measured on four plots: I, II, III
(Napa Valley) and IV (Graves). Each of the four plots
included one deep and one shallow soil location, identi®ed
as `D' and `S', respectively. Details of each plot are
presented in Table 1.
The top of the canopy on each plot was pruned
mechanically and identically within a plot. Within a plot,
Annals of Botany 87: 477±483, 2001
doi:10.1006/anbo.2000.1361, available online at http://www.idealibrary.com on
0305-7364/01/040477+07 $35.00/00 # 2001 Annals of Botany Company
* For correspondence E-mail: k-van-leeuwen@enitab.fr

2. the canopy was less dense on the shallow soil than on the
deep soil, but no quantitative measurements of leaf area
were taken.
From 6 June (day 157) to 10 October (day 284), no
signi®cant rainfall occurred in Yountville, California. Plot
IV (France) su€ered a drought from 6 June (day 157) to
27 July (day 208). On 27 and 28 July, heavy rainfalls
brought 25 and 29 mm precipitation, respectively.
Water potentials
DawnC, leafC and stemC were measured with a pressure
chamber equipped with a digital LCD manometer (SAM
Precis 2000, 33175 Gradignan, France). DawnC was
measured at the end of the night (between 2 h prior to,
and at, dawn) on uncovered leaves. LeafC was measured on
mature leaves which remained exposed to direct sunlight at
least 1 h before measurement. StemC was measured on
non-transpiring leaves that had been bagged with both
plastic sheet and aluminium foil at least 1 h before
measurement. Bagging prevented leaf transpiration, so
leaf water potential equalled stem water potential (Begg
and Turner, 1970). Leaves on the shaded side of the row
were selected for bagging to avoid overheating during the
bagging period. Bagging 1, 2, and 6 h prior to water
potential measurement gave the same results (data not
shown). Values of dawnC, leafC and stemC are the mean
of eight measurements collected on eight adjacent vines.
Measurement of midday stemC carried out simul-
taneously on a basal leaf and on an upper leaf (approx.
1 m above the basal leaf) on the same shoot showed an
average midday stemC shoot gradient lower than 0 to
0.05 MPa (data not shown). Thus, the stemC gradient of
pruned shoots was considered negligible according to Liu
et al. (1978).
On plots I, II and III (California), dawnC was measured
six times from 20 July (day 201, prior to veraison) to
5 October (day 278, harvest). LeafC and stemC measure-
ments were taken from 11 August (day 223) to 5 October.
Plots I and II were specially surveyed for seasonal
predawnC, leafC and stemC changes. Measurements were
taken on 11 August (diurnal in plots I and II), 27 August
(day 239) and 28 August (diurnal in plot I, midday values in
plots II and III), 4 September (day 247, diurnal in plot I,
midday values in plot II), 12 September (day 255, diurnal in
plot II), 19 September (day 262, diurnal in plot I, midday in
plot II) and 5 October (day 278, diurnal in plots I and II).
Midday VPD ranged from 2.8 to 4.3 kPa on these
measurement days.
On 27 September (day 270, on I-S, II-S and III-S) and
30 September (III-S) both midday leafC and stemC were
measured on six adjacent vines with four replicates per vine.
These measurements of leafC as well as stemC were
collected simultaneously on four di€erent shoots during
the steady period of the water potential diurnal curve (1100
to 1400 h). Forty-eight measurements were performed at
one location in 50 min. This experiment assessed the e€ect
of leaf and vine on both measurements.
In block IV (France), dawnC and midday stemC were
measured on the same vines throughout a long drought
period (7 weeks). As leaves were limited, leafC was only
measured on three occasions (beginning, middle and end of
the period). Midday VPD ranged from 2.9 to 3.6 kPa on
the measurement days.
Leaf transpiration
Leaf transpiration ¯ow (T) was measured with a steady
state porometer (LI-600, LI-COR, Lincoln, Nebraska
68504, USA). In plot IV, T, leafC and stemC were
measured simultaneously eight times on eight vines, on 17
and 24 July (days 198 and 205), from 1130 to 1350 h.
Midday air temperature and VPD on days 198 and 205 were
31 and 32 8C and 3.2 and 3.5 kPa, respectively.
DC was calculated as the di€erence between stemC
and leafC measured simultaneously on two mature
leaves on the same shoot. Following Begg and Turner
(1970), DC ˆ stemC À leafC ˆ rT, where r is resistance
to water ¯ow between the stem and a non-distal leaf, and T
is the mean transpiration ¯ow through the leaves of the
stem.
Resistance can be considered constant for mature leaves,
as it is determined by the radius of the xylem vessel and the
number of vascular bundles, so long as embolism does not
occur (Dimond, 1996; Tyree and Sperry, 1988; Schultz and
Matthews, 1993). Hence, an increase in DC indicates
increased transpiration ¯ow through the leaf, if a constant
petiole conductivity is assumed.
TABLE 1. Parameters of the various plots
Plot Location Soil texture
Root zone
depth (m) Variety Rootstock
Year of planting and
trellising system
Spacing (m) and
vines per ha
Height of the
leaf curtain (m)
I-D
I-S
Napa Valley
Yountville
Silty sandy clay
Sandy silty clay
4
2
Cabernet
Sauvignon
110 R 1990, Lyre trellis
double Guyot pruned
2.7 Â 1.5
2470
1.3 (for each
curtain)
II-D
II-S
Napa Valley
Yountville
Silty sandy clay
Sandy clay
2.5
1.5
Cabernet
Sauvignon
Vitis
rupestris L.
1972, Double Guyot
pruned
2.7 Â 1.5
2470
1.3
III-D
III-S
Napa valley
Yountville
Silty sandy clay
Silty sandy clay
3
2
Cabernet
Sauvignon
Vitis
rupestris L.
1990, Lyre trellis
double Guyot pruned
2.7 Â 1.5
2470
1.3 (for each
curtain)
IV-D
IV-S
Bordeaux area
Pessac Leognan
Sandy clay
Gravelly sandy
2
1.3
Cabernet
Sauvignon
SO4 1970, Double Guyot
pruned
1.1 Â 1
9090
0.8
478 Chone et al.ÐStemC is a Sensitive Indicator of Grapevine Water Status

3. Air humidity and temperature were recorded by
electronic meteorological stations located in the two
vineyards studied.
The statistical signi®cance of di€erences was veri®ed with
the Student's t-test means comparison and single factor
ANOVA (StatBox pro Software).
RESULTS AND DISCUSSION
Repeatability of C measurement in ®eld-grown vines
StemC and leafC were measured on the same vine using
four repetitions per vine for both measurements. The
standard error per vine was much lower for midday stemC
than for midday leafC (Table 2), hence, midday stemC
exhibited higher signi®cant vine to vine di€erences. In plum
trees (Shackel et al., 1997), stemC has been shown to be
more important in revealing tree to tree di€erences in water
status as soil conditions become drier. Midday leafC did
not indicate signi®cant vine to vine di€erences, or indicated
less signi®cant di€erences than midday stemC (Table 2). It
can be concluded that stemC is a better estimator of plant
water status than leafC. StemC di€erentiates better than
leafC due to the small variability between bagged leaves on
di€erent shoots of the same vine. LeafC is much more
variable, depending on the local climate. Grapevine is a
water stress avoiding species. Stomatal conductivity is
regulated to control leaf water de®cit, and maintains
midday leafC at a constant value (Naor, 1998).
Relationship between DC and T for grapevines
DC was signi®cantly (P 5 0.001) correlated with tran-
spiration ¯ow (Fig. 1) as was stemC (P 5 0.001) (Fig. 2).
Conversely, leafC was not correlated with transpiration
¯ow (Fig. 3). Hence, DC can be used to evaluate shoot
transpiration ¯ow. It has previously been proposed that DC
values of peach and plum trees are directly dependent on
leaf transpiration (Garnier and Berger, 1985). Here, the
relationship is experimentally established for vines (Vitis
vinifera). Naor and Wample (1994) observed the same
relationship in Vitis labruscana. The signi®cant correlation
of DC with transpiration ¯ow is mainly explained by
stemC, as leafC was not correlated with transpiration ¯ow.
LeafC was measured mainly at midday (1130 to 1350 h)
and was very much less variable than stemC due to
stomatal regulation of transpiration to maintain minimum
day leaf water potential and to prevent permanent leaf
damage. MacCutchan and Shackel (1992) found similar
results for plum trees, and concluded that a reduction in
transpiration and stomatal conductance was related to
stemC but not to leafC. In vines, Winkel and Rambal
TABLE 2. LeafC and StemC analysis of variance comparison, on plots I, II, III
Group
I-S
27 September
II-S
27 September
III-S
27 September
III-S
30 September
LeafC StemC LeafC StemC LeafC StemC LeafC StemC
Vine 1 MPa À1.21 À0.75 À1.37 À1.13 À1.34 À1.02 À1.29 À0.99
Vine 2 MPa À1.20 À0.79 À1.39 À1.08 À1.35 À0.85 À1.35 À0.89
Vine 3 MPa À1.22 À0.77 À1.37 À1.02 À1.32 À0.96 À1.18 À0.73
Vine 4 MPa À1.24 À0.81 À1.40 À0.99 À1.26 À0.72 À1.32 À0.78
Vine 5 MPa À1.18 À0.8 À1.39 À0.99 À1.3 À0.85 À1.21 À0.67
Vine 6 MPa À1.24 À0.75 À1.31 À0.68 À1.28 À0.93 À1.34 À0.93
F n.s. n.s. n.s.
signi®cance P 5 0.001 P 5 0.001 P 5 0.001 P 5 0.01 P 5 0.001
n.s., non signi®cant.
Data are means of four measurements collected on the same vines. Measurements were made between 1100 and 1300 h.
0.80.60.40.20
0
2
4
6
8
10
Transpirationflow(µgcm–2s–1)
DY (MPa)
y = 8.598x + 1.7577
R2 = 0.6548
FIG. 1. Relationship between DC and transpiration ¯ow.
–1.2 –1 –0.8 –0.4–0.6
Stem Y (MPa)
10
8
6
4
2
0
Transpirationflow(µgcm–2s–1)
y = 8.4201x + 10.603
R2 = 0.73
FIG. 2. Relationship between stemC and transpiration ¯ow.
Chone et al.ÐStemC is a Sensitive Indicator of Grapevine Water Status 479

4. (1993) have already observed that the same leaf water status
may correspond to very contrasting plant water relations.
In non-limiting light conditions, it is assumed that
stomata close and transpiration declines as water stress
develops (Hsiao, 1973). Given that during the experiment,
air temperature and VPD were identical for the vines
studied within plot IV, variation in stemC and transpiration
mainly re¯ected soil water availability. It is further
indicated that for a decrease in stemC of 0.5 MPa,
transpiration was reduced by about 60 %, which is similar
to values reported for plum trees by MacCutchan and
Shackel (1992). It can be concluded that stemC is a good
indicator of vine transpiration when measured close to the
sun zenith, as leafC reaches a constant minimum value.
Comparison of dawnC, stemC and leafC as indicators of
vine water status
Long-term water de®cit. From July to October, dawnC
remained constantly high on plots I-D, II-D and III-D.
Throughout the same period, dawnC decreased progress-
ively on II-S, whereas it only began to decrease after
1 September (day 244) on I-S and after 10 September
(day 253) on III-S, re¯ecting the occurrence of drying soil
conditions in these shallow root zones during the season
(Fig. 4). From 11 August (day 223) to 5 October (day 278)
on plots I, II and III, on both deep and shallow soil, dawnC
ranged from À0.1 to À0.72 MPa.
During the same period, midday leafC on plots I, II and
III ranged from À1 to À1.53 MPa, while midday stemC
varied from À0.45 to À1.4 MPa. Hence, of the three water
potential measurements, stemC exhibited the largest
amplitude.
On 11 August, di€erences in dawnC on plots I and II in
shallow and deep soil were small but statistically signi®cant,
whereas di€erences in leafC were not (Fig. 5A and B).
Conversely, on shallow soils, midday stemC was consist-
ently more negative than on deep soils, and the di€erences
were large and highly signi®cant (Fig. 5A and B). More-
over, in block III (28 August, day 240), only stemC was
statistically di€erent between III-S and III-D (Table 3).
–1.4 –1.2 –1 0.8
Leaf Y (MPa)
10
8
6
4
2
0
Transpirationflow(µgcm–2s–1)
y = 3.744x + 8.217
R2 = 0.039
FIG. 3. Relationship between leafC and transpiration ¯ow.
Day of year
272242212182
0.0
–0.2
–0.4
–0.6
–0.8
PredawnleafY(MPa)
I-S
I-D
II-S
II-D
III-S
III-D
FIG. 4. Seasonal dawnC on the six plots located in California. Error
bars indicate s.e.
180015001200090006000300
Time of day (h)
180015001200090006000300
–1.5
–1.2
–0.9
–0.6
–0.3
0
0
–0.3
–0.6
–0.9
–1.5
–1.2
Y(MPa)Y(MPa)
A
B
I-D stem water potential
I-S stem water potential
I-D leaf water potential
I-S leaf water potential
II-D stem water potential
II-S stem water potential
II-D leaf water potential
II-S leaf water potential
FIG. 5. StemC and leafC diurnals on plot I and plot II (11 Aug. 1998).
Data are means of eight measurements collected on eight adjacent
vines. Error bars indicate s.e.
480 Chone et al.ÐStemC is a Sensitive Indicator of Grapevine Water Status

5. This suggests that stemC is a more sensitive indicator of
nascent water limitation than dawnC.
From September to 5 October, on plots I and II, leafC
di€erences became statistically signi®cant between shallow
and deep soils, due to increasing water uptake limitation on
I-S and II-S. Yet stemC di€erences remained greater than
leafC di€erences (Fig. 6A and B). DawnC di€erences
between shallow and deep soils occurred earlier than
leafC di€erences (Fig. 5A and B). As expected, dawnC
appeared to be more discriminating than leafC.
PredawnC and stemC evolution between two periods of
rainfall under the Bordeaux climate
In early June the soil on plot IV was considered to be at
®eld capacity after winter and spring precipitation. Through-
out a period without signi®cant rainfall (from early June to
27 July, day 208) midday stemC indicated signi®cant
di€erences between IV-S and IV-D on three occasions,
whereas dawnC did not (Fig. 7). When the water de®cit
became more severe, both indicators exhibited signi®cant
di€erences. Lastly, on 30 July (day 211), after heavy rainfall,
only midday stemC continued to show a signi®cant
di€erence between the two blocks. Throughout this period,
midday leafC was not signi®cantly di€erent between IV-S
and IV-D (data not shown). As stemC revealed water status
di€erences much before dawnC, it is con®rmed to be a more
sensitive indicator than dawnC. Of the three pressure
chamber applications, the order of sensitivity to developing
water de®cit is: stemC 4 dawnC 4 leafC.
After rainfall on 27 and 28 July (day 208 and 209, dawnC
rapidly recovered and did not di€erentiate between soil
depths. Conversely, midday stemC still revealed di€erences
in water status of grapevines. Under mild water de®cit, the
vine obtains sucient water from the soil overnight for its
water status to have recovered fully by dawn, whereas at
midday bulk leaf transpiration exceeds root water uptake
capacity. This short-term de®cit (several hours around
midday) can only be revealed by stemC. This is a crucial
point in viticulture since Schultz and Matthews (1988)
observed that cavitation of the xylem shoot apex could
restrain vegetative growth even at moderate water de®cits.
Early shoot growth slackening is related to good fruit
ripening conditions (Van Leeuwen and Seguin, 1994).
Seasonal variation in transpiration in a non-irrigated
vineyard (California)
DC was used to compare grapevine transpiration in the
di€erent plots. On Plot II (Figs 8 and 9), DC values on
shallow soil were lower than on deep soil. Furthermore, on
A
B
272242212
Day of year
–1.8
–1.5
0.0
–0.3
–0.6
–0.9
–1.2
–0.0
–0.3
–0.6
–0.9
–1.2
–1.5
–1.8
Y(MPa)Y(MPa)
II-S stem water potential
II-D stem water potential
II-S leaf water potential
II-D leaf water potential
I-D stem water potential
I-D leaf water potential
I-S leaf water potential
I-S stem water potential
FIG. 6. Seasonal midday leafC and stemC on plots I and II (Napa).
Data are means of eight measurements collected on eight adjacent
vines. Error bars indicate s.e.
TABLE 3. Plot III, 28 August, dawnC, leafC and stemC
DawnC
(0400 h)
leafC
(1300 h)
StemC
(1315 h)
III-S À0.12a À1.2a À0.8a
III-D À0.12a À1.3a À0.6b
Data are means of eight measurements. Plots III-S and III-D were
compared. Means within a column followed by the same superscript
are not statistically di€erent at P ˆ 0.05
213182151
Day of year
Y(MPa)
0
–0.3
–0.6
–0.9
–1.2
IV-S stem water potential
IV-D stem water potential
IV-S predawn water potential
IV-D predawn water potential
FIG. 7. Seasonal dawnC and midday stemC through a drought period
(plot IV, Bordeaux).
Chone et al.ÐStemC is a Sensitive Indicator of Grapevine Water Status 481

6. 12 September and 5 October (day 255 and day 278) in II-S,
DC (diurnal) began to reach a plateau or decreased before
noon, whereas in II-D, DC increased until noon. This
suggests reduced transpiration through stomatal closure on
II-S. Di€erences in transpiration between II-S and II-D
re¯ected only di€erences in soil water availability, as
evaporative demand and canopy size were similar on
shallow and deep sites of one plot during the same day.
These observations are consistent with research done in
citrus trees (Cohen et al., 1983) and in peach trees (Garnier
and Berger, 1985).
In II-S, DC curves obtained with data collected on
12 September (Fig. 8) and 5 October (Fig. 9) re¯ected two
phases of stomatal regulation during the day. During the
®rst hours after dawn, DC was signi®cantly lower on II-S
than II-D, suggesting moderate transpiration regulation on
II-S. From 1000 h (12 September) and from 0730 h
(5 October) II-S DC decreased or stabilized, whereas in
II-D, DC was still increasing (following the developing
evaporative demand). This re¯ected intense transpiration
regulation on II-S. Similar responses were obtained on plot
I (data not shown).
DC measured on the two types of soil in plots I and II
showed that transpiration rate is correlated with soil depth,
and probably to available water. This di€erence in
transpiration between the shallow and deep root zone can
be related both to a regulation of stomatal conductance and
to leaf area per shoot. Whatever the reason, it can be
concluded that grapevines have the capacity to adjust their
water use to soil water availability.
Daily C gradient
Assuming that dawnC represents soil water potential in
the root zone (Begg and Turner 1970), the dawnC-stemC
gradient was determined to compare it with the stemC-
leafC gradient in the same SPAC. On 11 August (Table 4),
under conditions of unlimited water uptake (I-D), the
dawnC-stemC gradient accounted for 36 % of the total
dawnC-leafC gradient, whereas in conditions of limited
water uptake (I-S), the dawnC-stemC gradient accounted
for 55 % of the total dawnC-leafC gradient. As demon-
strated in this paper, the lower stemC-leafC gradient on I-S
re¯ected reduced transpiration. A low stemC-leafC gradi-
ent resulted in a high dawnC-stemC gradient, re¯ecting
reduced hydraulic conductivity of the soil-stem pathway.
Conversely, a low dawnC-stemC gradient on I-D was
related to high hydraulic conductivity in the plant.
A comparison of the dawnC-stemC gradient in I-D and
I-S under the same evaporative demand indicated that the
gradient re¯ects vine water uptake conditions. As observed
previously in this paper, when vine water de®cit occurs in
shallow soil, dawnC di€erences between shallow and deep
plots are small or non-existent compared to stemC
di€erences. In these conditions the dawnC-stemC gradient
is essentially determined by stemC. Thus, these results
suggest that in vines stemC corresponds to hydraulic
conductivity within the vine trunk and canes from roots
to petioles. This deduction is consistent with the result of
Tyree and Ewers (1991) who showed that lianas have high
speci®c conductivity but low Huber value compared with
other angiosperms. The Huber value is de®ned as the
sapwood cross-section divided by the leaf area distal to the
stem. Speci®c conductivity is hydraulic conductivity divided
by sap wood cross-section. The high speci®c conductivity of
the vine trunk and shoot xylem pathway means low sap
¯ow resistance within this pathway. Liu et al. (1978)
180015001200090006000300
0
0.2
0.4
0.6
0.8
DY(MPa)
Time of day (h)
DY II-D
DY II-S
FIG. 8. DC diurnals on plot II (12 Sep. 1998).
0
0.2
0.4
0.6
0.8
DY(MPa)
180015001200090006000300
Time of day (h)
DY II-D
DY II-S
FIG. 9. DC diurnals on plot II (5 Oct. 1998).
TABLE 4. DawnC- midday stemC and midday stemC-
midday leafC gradient proportions in the total dawnC-
midday leafC gradient on Plot I (11 Aug. 1998)
Location I-D I-S
DawnC-stemC (MPa) 0.4 0.67
StemC-leafC (MPa) 0.71 0.54
DawnC-leafC (MPa) 1.11 1.21
((dawnC-stemC)/(dawnC-leafC)) Â 100 ( %) 36 55
482 Chone et al.ÐStemC is a Sensitive Indicator of Grapevine Water Status

7. reported that for cultivated Vitis labrusca L. plants, the
total resistance of the stem was much lower than that of the
roots and leaves.
This demonstration further justi®es the use of stemC as a
comprehensive indicator of both vine water uptake
conditions and vine transpiration.
CONCLUSIONS
StemC is demonstrated to be a comprehensive indicator of
early water de®cit in plants, while leafC is not. Within the
limits of the experimental conditions of this study, we
observed that leafC regulation was independent of soil
moisture supply. When ®eld-grown grapevines experience
water limitation, the ®rst indicator of this water de®cit is
midday stemC, followed by dawnC. Midday leaf water
potential is a less signi®cant indicator of water constraint.
In this research, signi®cant variation in dawnC was
related to soil type. This study also demonstrated that DC
(stemC À leafC) is signi®cantly correlated to transpiration
¯ow and thus can be a valuable method for determining
®eld-grown vine transpiration. It was further con®rmed that
in non-irrigated ®eld-grown vines the midday stemC was
highly correlated to transpiration whereas midday leafC
was not. In discussing dawnC-stemC and stemC-leafC
gradients, we conclude that stemC is an indicator of
hydraulic conductivity in the trunk and shoot sap pathway.
Mild to moderate water de®cits have positive oenological
e€ects on berry development and ripening. Therefore,
stemC appears to be a powerful tool to assess vine water
status and to manage it in the vineyard. This study
demonstrated that stemC is such an indicator for non-
irrigated vines. StemC should also be an accurate indicator
for vine irrigation management.
ACKNOWLEDGEMENTS
Assistance of Dr K. Shackel in the ®rst approach of stem
water potential was invaluable and opportune. We thank
C. and C. Moueix (owners of Dominus Estate), for their
®nancial support and encouragement. We also thank
J. C. Berrouet, J. M. MaureÁ ze, B. Champy, J. Delmas,
J. P. Masclef and P. Baratie .
LITERATURE CITED
Begg JE, Turner NC. 1970. Water potential gradients in ®eld tobacco.
Plant Physiology 46: 343±346.
Cohen Y, Fuchs M, Cohen S. 1983. Resistance to water uptake in
mature citrus tree. Journal of Experimental Botany 34: 451±460.
Dimond AE. 1966. Pressure and ¯ow relations in vascular bundles of
the tomato plant. Plant Physiology 41: 119±131.
Garnier E, Berger A. 1985. Testing water potential in peach trees as an
indicator of water stress. Journal of Horticultural Science 60:
47±56.
Greenspan M, Schultz H, Matthews MA. 1996. Field evaluation of
water transport in grape berries during water de®cits. Physiologia
Plantarum 97: 55±62.
Hsiao TC. 1973. Plant responses to water stress. Annual Review of Plant
Physiology 24: 519±570.
Liu WT, Wenkart W, Allen LH, Lemon ER. 1978. Soil-plant-water
relations in a New York vineyard: resistances to water movement.
Journal of American Society of Horticulture and Science 103:
226±230.
MacCutchan H, Shackel KA. 1992. Stem water potential as a sensitive
indicator of water stress in prune trees. Journal of American
Society of Horticulture and Science 117: 607±611.
Matthews MA, Ishii R, Anderson MM, O'Mahony M. 1990.
Dependence of wine sensory attributes on vine water status.
Journal of Science of Food and Agriculture 51: 321±335.
Naor A. 1998. Relations between leaf and stem water potentials and
stomatal conductance in three ®eld-grown woody species. Journal
of Horticultural Science and Biotechnology 73: 431±436.
Naor A, Wample RL. 1994. Gas exchange and water relations of ®eld-
grown Concord (Vitis labruscana Bailey) grapevines. American
Journal of Enology and Viticulture 45: 333±337.
Scholander PF, Hammel HJ, Bradstreet A, Hemmingsen EA. 1965. Sap
pressure in vascular plants. Science 148: 339±346.
Schultz HR, Matthews MA. 1988. Resistance to water transport in
shoots of Vitis vinifera L. Plant Physiology 88: 718±724.
Schultz HR, Matthews MA. 1993. Xylem development and hydraulic
conductance in sun and shade shoots of grapevine (Vitis vinifera
L.), evidence that low light uncouples water transport capacity
from leaf area. Planta 190: 393±406.
Shackel KA, Ahmadi H, Biasi W. 1997. Plant water status as an index
of irrigation need in deciduous fruit trees. Hortechnology 7: 23±29.
Tyree MT, Ewers FW. 1991. The hydraulic architecture of trees and
other woody plants. New Phytologist 119: 345±360.
Tyree MT, Sperry JS. 1988. Do woody plants operate near the point of
catastrophic xylem dysfunction caused by dynamic water stress?.
Plant Physiology 88: 574±580.
Van Leeuwen C, Seguin G. 1994. Incidences de l'alimentation en eau de
la vigne, appre cie e par l'e tat hydrique du feuillage, sur le
de veloppement de l'appariel ve ge tatif et la maturation du raisin.
Journal International des Sciences de la Vigne et du Vin 28: 81±110.
Winkel T, Rambal S. 1993. In¯uence of water stress on grapevines
growing in the ®eld: from leaf to whole-plant response. Australian
Journal of Plant Physiology 20: 143±157.
Chone et al.ÐStemC is a Sensitive Indicator of Grapevine Water Status 483