l(lurnal of Horticultural Science (1989) 64 (5) 541-552

Response of peach tree growth and cropping to soil water

deficit at various phenological stages of fruit development
By S.-H. LI,J* J.-G. HUGUET,l P. G. SCHOCHJ and P. ORLAND<Y
JI.N.R.A.-S.R.I.V., Domaine de Gotheron, 26320 St Marcelles Valence, France
zI.N.R.A., Station d' Agronomie, Centre de Recherche Agronomique d' Avignon, 84140
Montfavet, France
JI.N.R.A., Station de Bioclimatologie, Centre de Recherche Agronomique d'Avignon, 84140
Montfavet, France

SUMMARY
Four- and five-year-old 'Merrill Sunda~ce' peach trees, protected from rainfall by poly-
ethylene film covers, were fully irrigated using micro-sprinkler (irrigatio~ scheduling
based on a tensiometer technique), or subjected to water stress at different phenological
stages of fruit growth. Water deficit imposed during the first phase of rapid growth
significantly increased fruit size at harvest during two experimental years when compared
with the control full-irrigation treatment, while smaller fruits were produced from trees
receiving an imposed water deficit during the final accelerated fruit growth, or throughout
the fruit development period. When water deficit was applied to the trees during the pit
hardening phase and the first two phases of fruit development, fruit size was not affected.
However, shoot extension growth and limb diameter increases were limited whenever
water supply was restricted. After-effects on limb expansion growth and benefits of water
stress on fruit growth were obvious during the post-stress period. Moreover, premature
fruit drops after the June-drop were reduced for all the water stress treatments. The level
of total soluble solids was higher in fruits from the trees subjected to water stress during
the final rapid phase of fruit growth, and flower bud production was improved on trees
given a restricted supply of water during the critical period of flower bud induction.

IT is widely acknowledged that field crops such
as wheat, maize, sorghum and pea are more
sensitive to water stress at particular stages of
development (Begg and Turner, 1976; Grignac,
1985). At these critical stages, for example at
flowering for peas (Hiler et aI., 1972) and at
grain swell for sorghums (Langlet, 1973), a
restricted water supply reduces crop yield most
~
severely.
With fruit trees, there is little quantitative
information on their cmpping responses to
water stress at different phenological stages.
Chalmers et al. (1984) concluded from previous
experiments on peach and pear trees that
reduced water supply during the early stages of
fruit growth until the end of shoot growth, did
not affect final fruit size, number or yield.
.Permanent address: Horticulture Department, Huazhong
Agricultural University, Wuhan, People's Republic of
China.
Nevertheless, Vidaud et al. (1987) considered
the period from the beginning of fruit pit hard-
ening until the end of shoot growth in peach
trees to be a critical phase for fruit size and yield
responses to water supply. The effects of water
stress during the final stage of rapid fruit growth
on final fruit size have been reported as either
of little importance (Irving and Drost, 1987) or
remarkable (LOtter et aI., 1985).
All these investigations on fruit trees were
conducted under natural condition. Since the
water requirements of fruit trees are relatively
low during the first and second phases of fruit
development (Li et al., 1989a), a little rainfall
during these periods can remove the water
.stress status in trees. Therefore, a strict experi-
mental control of soil water status is required to
determine the effects of water deficit on fruit
trees at different phenological stages.
This paper reports an experiment designed to

.o1t
- ------
-. --~
-
542 Soil water deficits and peach
elucidate the behaviour of peach trees growing
in the field under water stress at different phe-
nological stages, carried out under the protec~
tion of polyethylene film covers for the
experimental plot. We studied.in particular:
first, tree growth, such as shoot extension and
stem diameter expansion, and fruit growth pat-
terns, secondly, fruit bud production, fruit drop
and yield and thirdly, fruit quality and fruit
storage.
MATERIALS AND METHODS
Experimental site and plant materials
This study was conducted during the growing
season of 1987 and 1988 in the, Gotheron
experimental orchard ofthe Institut National de
la Recherche Agronomique near Valence in the
middle Rhone Valley of France. The orchard
soil was stony alluvial with the mechanical com-
position of 15% clay, 30% silt and 54% sand
after removing the stones. Field capacity was
about 17% and wilting point about 8% by
weight.
The materials selected for this trial were pea-
ches, cv. Merrill Sundance (Prunus persica (L.)
Batsch), a late ripening cultivar on 'Rubira'
rootstock planted in the spring of 1984. The
FIG. 1
Covers protecting the experimental plot from rainfall. The arrow indicates the junction of two polyethylene films.
<It
trees, trained in a double Y conformation, we
spaced 3.5 m apart within the rows and 5
apart between rows.
Irrigation treatments
A 0.1 ha surface area of the orchard was PI'
tected from rainfall from 24 April 1987 (8
after full bloom) to 22 September 1988 (aft
the harvest) by using covers of black pol
ethylene films (6 or 8 m width and 200 J.l.mthic
ness) (Figure 1). Th~ ridges under the tr
were about 2.2 m wide, supported on a fr
about 0.6 m high which consists of fiat ir
arches (gauge: 5xl4mm) and 'Proven
reec!s: the ends of the arches (0.1? cm Ion
were pushed into the ground and all the arch
were bound toge~her with seven parallel 1'0
of 'Provence' reeds aligned along the. tre~ 1'0.
Water could thus be supplied by micro-spri
klers (two per tree) which wetted an area abo
1.2 m in diameter at a discharge rate of 20 I h-
The junctions of two polyethylene films.
which 'Provence' reeds were:wrapped were ti ,
to the arches, one:Ove1't.Qeother, and the abo
film was cut to accommodate the tree tru
and also to permit control of water applicatio
Six treatments were applied respectively
 S.-H. LI, J.-G. HUGUET, P. G. SCHOCHand P. ORLANDO
six rows of 7 to 10 trees, selected for uniformity
of the trunk girths and tree crown size (Table I).
No guard rows were used and a root-cutting
treatment was carried out, down t050 em, half
way between the rows, before the film cover
was installed, to limit crossed-root effects. The
six treatments applied were as follows:
A: control. Trees were irrigated whenever
the mean of the soil water potential, derived
from the readings" of three tensiometers
installed at 0.5 m depth and 0.5 m from the
emitters, reached. about -60 kPa. The soil
water potential usually rose to about -lO,kPa
after irrigation. The first irrigation wasapplled
on 8 May 1987 andon 3 May 1988;
B: water stress during the firs.t phase of fruit
development (1st PWS). Trees did not receive
any water. during the first stage of rapid fruit
growth (for description of the fruit develop-
ment phases, see the reviews of Bollard (1970)
and Romani and Jennings (1971», and were
then irrigated as in treatment A;
C: water stress during the second phase of fruit
development (2nd PWS). Irrigation occurred at
one-third the frequency of treatment A during
the stage of fruit pit hardening. For the remain-
ing time, the irrigation scheduling was as
applied in treatment A;
D: water stress during the third phase of fruit
development (3rd PWS). Irrigation was applied
at one-third the frequency of treatment A . was measured at the end of each phase in 1987,
during the final stage of rapid fruit growth. For
the remaining time, the irrigation scheduling
was as applied in treatment A;
E: water stress during the first two phases of
fruit development (1+ 2 PWS). Irrigation sched-
uling was the same as in treatment B for the first
TABLEI
Trunk girths and tree crown sizes before the differential
irrigation treatments were started
Trunk gi{th Tree crown size
111
Irrigation (cm) (m3)Y
Control 25.1 :t 0.& 5.8 :t 0.3
1st PWS 26.2 :t 0.5 6.3 :t 0.4
2nd PWS 25.5 :t 0.9 6.1 :to.4
3rd PWS 25.3 :t 0.8 5.7:t 0.2
1+ 2 PWS 26.6 :t 1.0 6.2 :t 0.2
1+2+3 PWS 25.4 :t 0.5 5.3 :t 0.4
.y
Average followed by standard deviation.
Tree crown size (V) estimated according to:
V = (D2 x H)/8
where D is the mean crown diameter (average of S-N diam-
eter and E- W diameter), H the crown height (Zhang et af. , a honeycomb, per tree was examined at inter-
1979).
543
phase, as in treatment C for the second phase,
and as in treatment A for the last phase of fruit
development;
F: water stress during the three phases of fruit
development (1+2+3 PWS). Irrigation sched-
uling was as in treatmentB for the first phase, as
in treatment C for the second phase, and as in
treatment D for the last phase of fruit
development.
The volume of water applied during the two
years at various stages for the six treatments
and the dates of the start and end of the three
phases are given in Table II.
Measurements
Eight shoots of the previous season per tree
were marked in both years at the 'beginning of
the growing season. Their flower numbers and
subsequent fruitlets borne were controlled to
estimate the stress effects on June drop. After
physiological drop of fruits, peaches were hand
thinne<ilO to 15 em apart on 22 June in 1987
and on 13 June in 1988, i.e. four or five peaches
were left on the long shoots (Vidaud and
Clanet, 1978; Mitcham, 1980). The remaining
fruit lets after hand thinning and fruits at har-
vest were counted for each tree, allowing the
premature fruit drop to be determined.
The total growth increment of all new shoots
on the previous season's shoots marked above
but only the length of the terminal new shoots
was measured in 1988. The rate of stem diam-
eter expansion was assessed by dendrometer
measurement on two limbs per treatment in
1987 and by the hand measurement on one limb
per tree in 1988. The dendrometer system used
linear variable displacement transducer
(Huguet, 1985) mounted in an INVAR frame
as described by Li et al. (1989a). Seven to ten
tagged fruits on. two previous season's shoots
per tree were usually measured every week to
follow the fruit growth pattern. Measurements
were made on the peach suture between 0600
and 0800 hours until harvest time.
All the fruits from each tree were weighed at
harvest. On the first picking date, percentage
soluble solids were determined on ten fruits per
tree with a refractometer. Moreover, a box of
18 selected peaches, placed in the paper cells of
vals for rot under home conditions.
 544 Soil water deficits and peach

TABLEII treatments. Thus, the water stress imposed

Volume of irrigation water applied to 'Merrill Sundance'
peach trees during the different phases of fruit growth during the third phase of fruit development did
not affect the shoot elongation of 'Merrill Sun-
Water applied (m3/ha)
during fruit growth phase dance' peach trees. Even in the second year, the
Irrigation treatment I II III potential of shoot growth during the first phase
1987
was very similar between the control treatment
Control 553 949 2069 and 3rd PWS. At the end of phase I, an average
1st PWS 0 949 2069 of 41.5 cm shoot growth increment was.
2nd PWS 553 388 2069
3rd PWS 553 949 895 recorded for the control trees in 1988compared
1+2 PWS 0 388 2069 with a 38.8 cm shoot growth increment for trees
1+2+3 PWS 0 388 895 of the treatment 3rd PWS.
Phases: start 8 May 27 June 7 Aug.
end 26 June 6 Aug. 24 Sept.
1988 Limb expansion growth
.
Control 414 1059 1816 All the water stress treatments strongly
1st PWS 0 1059 1816 reduced final stem diameter increment in both
2nd PWS 414 364 1816
3rd PWS 414 1059 620 years if compared with the control trees (Figure
1+2 PWS 0 364 1816 2). Limb expansion growth under 2nd PWS and
1+2+3 PWS 0 364 620
Phases: start 3 May 29 June 10 Aug.
'3rd PWS treatments was limited as soon as
end 28 June 9 Aug. 22 Sept. water stress was imposed on the trees. More-
over, the final reduction of limb diameter incre-
ment by restricted water supply was found to
Flower bud density data were obtained by depend on the time at which water stress was
sampling flower bud number on eight shoots, 30 applied to trees, but not on the duration of
to 60 cm long, per tree only in December 1987. water stress or on the restricted amount of
water finally applied.
Statistical analysis All trees which had suffered water stress
The replicated blocks in the present study during the first phase (treatments 1st PWS, 1+2
consisted of 7-10 single trees of each irrigation PWS and 1+2+3 PWS) had a much smaller
treatment. Due to the absence of a randomized increment of limb diameter than those water-
block arrangement, the method of one-way stressed during the second or third phase.
analysis of variance (Dagnelie, 1975; Zhang et The patterns of limb expansion growth were
al., 1979) was used in the statistical analysis of
these data. TABLE III
Effects of water deficit imposed during different phenological
stages on the shoot extension growth of cv. Merrill Sun dance
RESULTS peach trees, expressed as the extension growth increment of
Shoot extension growth all the new shoots per previous season's shoot in 1987 and the
extension growth increment of terminal shoot in 1988
Water stress imposed during the first and
second phase of fruit development respectively Shoot groWth increment (em) during
Irrigation treatment phase I phase II
resulted in a significant reduction of shoot
extension growth during the stress period in 1987
,. both years, as shown in Table III. The total Control 184.7 a' 81.8 ab
1st PWS 133,0 b 99.7 a
extension growth increment was reduced by 2nd PWS 191.8 a 61.0c
about 25% in 1987 (all the new shoots on the 3rd PWS 212.8 a 82,2 ab
previous season's shoot) and from 11 to 40% in 1+2 PWS 153.0 b 68.4 be
1+2+3 PWS 148.9 b 67.3 be
1988 (terminal shoots). However, there were
1988
no obvious differences in shoot growth during Control 41.5 a 16.6 ab
the second phase between the treatments 2nd 1st PWS 33.6c 13.4 be
PWS and 1+2 PWS, nor between the control 2nd PWS 39.6ab 11.0 cd
3rd PWS 38.8 ab 17.2 a
trees and the rewatered trees (e.g. treatment 1+2 PWS 35.8bc 9.9d
1st PWS). 1+2+3 PWS 36.9bc 12.3 cd
Shoot extension growth stopped in mid-July 'Figures followed by different letters are significantly dif-
or at the end of July for the trees of all the ferent at P = 0.05.

j
~
j
 S.-H. LI, J.-G. HUGUET, P. G. SCHOCH and P. ORLANDO 545

20

E 16
E
w 12 PHASE I
~
~
(J
8
z
4 PHASE III
0::
w
tA 0
::E
«
Q 12
"PHASE I
In 8
::E
::i
4 PHASE III

0
May June July Aug. Sept.
FIG. 2
Limb growth of 'Merrill Sundance' peach trees in 1987 (top) and in 1988
(bottom) in response to water deficit applied during different phenological
phases of fruit development: Control (8--.), 1st PWS treatment
(0---0), 2nd PWS treatment (A-A), 3rd PWS treatment
(6---6), 1+2 PWS treatment (8-8) and 1+ 2+3 PWS treatment
(0---0).

very similar between the trees of treatments 1st
PWS, 1+2 PWS and 1+2+3 PWS, with no
marked differences in final increment of limb
growth, except that ofthe 1st PWS treatment in
1988. For trees in this treatment, limb diameter
increased more rapidly during the last phase
than those of the other two treatments.
Trees in the 3rd PWS treatment, with a
reduced amount of water of 1174 m3ha-1
applied in 1987 and of 1196m3ha-1 in 1988, had
a greater increment of final limb diameter than
those of the 1st PWS treatment with a reduced
water volume of 553 m3ha-I in 1987 and 414
m3ha-I in 1988.
Fruit growth and fruit size at harvest
Seasonal fruit growth of 'Merrill Sundance'
peaches shows a classical pattern of three dis-
tinct growth periods (Figure 3).
In 1987, water stress imposed on trees did not
influence fruit growth until the end of the
second phenological phase: there were no sig-
. nificant differences in fruit size between all the
treatments before 1 August. On 7 August,
fruits on the trees of treatments 2nd PWS and
1+ 2 PWS were significantly smaller than those
on the control trees. Unlike the effect of pre-
vious treatments, fruits of the 1st PWS treat-
ment were larger than the control ones. Water
stress imposed during the phase III limited fruit
growth, resulting in a significantly smaller fruit
size in the treatment 3rd PWS from 2 Septem-
ber. However, ftuit size in the treatments 2nd
PWS and 1+ 2 PWS became similar to that in
the controls about 1 and two weeks after remov-
ing the water stress. This similarity in size was
then maintained until maturity.
In 1988, treatment effects on fruit develop-
ment were similar to those in 1987, except with
'lst PWS. From early July, the fruit diameter on
the trees of this latter treatment was signifi-
cantly greater than that of the other treatments,
including that of the control treatment.
At harvest, fruit size was considerably
 546 Soil water deficits and peach

70 NS :I:
---------------------------- I

60
PHASE I PHASE II

E 50
E

w 40
~
w
5z 30 PHASE III

20
~ =r
I I
~ 70 !l_S I I I I I I
:IE
c:(
Q
i
I
I- /..
P
60 I- I cf~B
PHASE I i PHASE II
-
I ,. ~~
".7,~
1
i /~
I
!:so
;:)
!
I. i,.P/ """"
,./
0:: I%;,. /. ,/
u..
! /1~a
40 I
i _o-~---
~1<'
PHASE 1/1

~
I

I I
30 """,- I i
.,- i I
I I I i.
May June July Aug. Sept.
,
FIG. 3
Fruit growth of 'Merrill Sundance' peaches in 1987 (top) and in 1988 (bottom)
in response to water deficit applied during different phenological phases of fruit
development: Control (8-8), 1st PWS treatment (0---0), 2nd
PWS treatment (A-A), 3rd PWS treatment (6---6), 1+2 PWS
treatment (8-8) and 1+2+3 PWS treatment (0---0). Vertical
bars represent L.S.D. at P = 0.05.

improved by the treatment 1st PWS, an
increase of 12 g in both years, when compared
with that of the control treatment (Table IV).
By contrast, significantly smaller fruits were
obtained on the trees of the treatments 3rd
PWS and 1+2+3 PWS. A decrease of 1~14 g
in 1987 and 12-15 g resulted from the two pre-
vious water-stress treatments. As in treatments
2nd PWS and 1+ 2 PWS, water deficit had little
effect on the final fruit fresh weight.
Fruit-set and crop yield
Effects of water deficit applied at different
phenological stages on fruit-set were evaluated
by studying fruit drops of June and pre-harv
(Table IV).
Water stress imposed on peach trees did n
enhance the"ir June-drop. In 1987, fruit-set
water-stressed trees (treatments 1st PWS, 1
PWS and 1+2+3 PWS) was 20.6%, a val
similar to that with normally irrigated tr
(mean of control, 2nd PWS and 3rd PWS tre
ments). In 1988, the extent of fruit drop in
water-stressed trees was similar or less im
tant than that of the control trees, although'
significant differences existed.
However, water deficit at every stage
during all stages of fruit development improv

L
 S.-H. LI, J.-G. HUGUET, P. G. SCHOCH and P. ORLANDO 547

TABLEIV
Influence of water deficit imposed during different phenological stages of fruit development on the cropping characteristics of cv.
Merrill Sundance peaches
Fruit Fruit drop Mean Fruit bud
June fruit number prior to Fruit fruit number per
Irrigation drop' after maturity' production weight m shoot
treatment ('Yo) thinning ('Yo) (kg/tree) (g) length
1987
Control 75.5b' 122.7 a 16.4 a 16.71 ab 164.1 b 45.8b
1st PWS 77.3 b 120.0 a 7.6b 19.64 a 176.1 a 48.3 ab
2nd PWS 76.1 b 122.0 a 5.3b 19.39 a 167.5 b SO.3a
I 3rd PWS 85.0 a 92.5b 5.0b 13.68 b 154.0 c 48.0 ab
1+ 2 PWS 77.5 b 124.0 a 7.5 b 19.01 a 165.9 b 49.9 a
1+2+3 PWS 83.3 a 85.1 b 5.0b 12.16 b 150.5 c 49.7 a
1988
Control 60.0 370.0 6.3 43.75 126.2 b
1st PWS 60.2 350.0 6.8 45.16. 138.5 a
2nd PWS 51.6 383.0 *' 3.4 44.84 121.0 be
3rd PWS 46.9 380.9 4.7 42.18 116.2 cd
1+2 PWS 57.2 353.8 4.5 42.00 . 124.3 bc
1+2+3 PWS 55.7 326.0 2.2 35.35 110.9 d
,
, From
From
full bloom to the end of physiological drop.
hand thinning after fruit physiological drop to harvest.
,
Figures followed by different letters are significantly different at P = 0.05.

fruit set after June drop in 1987. A significantly ments, because of a lighter load of total fruitlets
more important fruit drop after the June phys- after hand thinning (Table IV) resulting from
iological drop to maturity was found with the the greater June fruit drops, and smaller fruit.
control trees than the water-stressed ones. This In 1988, there were no significant differences in
effect on premature fruit drop disappeared in yield. However, the trees of the 1+2+3 PWS
1988. For this second year, few fruits dropped treatment produced about 16-22% fewer fruits
prior to maturity on all the experimental trees, compared with the others.
including the control ones, and no significant
differences were observed. Fruit quality
In 1987, fruit drop and fruit size were Trees subjected to water stress only during
reflected in crop yield. A yield increase of 14 to the third phase (treatments 3rd PWS and
18% was recorded on the trees under the treat- 1+2+3 PWS) produced fruits with higher total
ments 1st PWS, 2nd PWS and 1+2 PWS when soluble solids at harvest in both years than did
compared with the fruit production of the con- the control trees (Table V). However, water
trol trees, but this improvement was statis- stress imposed to trees during phase I or II, or
tically significant only at about P = 0.10. In phases I and II had no obvious effect on the
addition, a lower yield, a decrease of 18 to 27% content of soluble solids in the fruits.
compared with the control, was obtained on the Water stress treatment applied during the
, trees under 3rd PWS and 1+2+3 PWS treat- third phase improved fruit keeping capacity
\I TABLE V
Content of the total soluble solids of cv. Merrill Sundance
TABLEVI
Rate of flower-induced buds (flower buds/total buds) of cv.
peach fruits (first picking) related to irrigation treatment Merrill Sundance peach trees in 1987 related to irrigation
I Soluble solids ('Yo)
treatment
Irrigation treatment in 1987 in 1988 Irrigation treatment Rate of flower-induced buds ('Yo)
Control 12.8 be' 13.2 be Control 51.6 c'
1st PWS 12.5 c 13.6 ab 1st PWS 56.2 be
2nd PWS 13.0 ab 13.1 c '2nd PWS 59.7 ab
3rd PWS 13.2 a 14.1 a 3rd PWS 55.0 be
1+2 PWS 12.8 be 13.3 be 1+2 PWS 59.6 ab
1+2+3 PWS 13.2 a 13.7 ab 1+2+3 PWS 61.5 a
,
Figures followed by different letters are significantly dif- 'Figures followed by different letters are significantly dif-
ferent at P
= 0.05. ferent at P = 0.05.

.
548 Soil water deficits and peach

I I I
NS
70
I
60
I I
(/)

t:
~
&...
50
0
,-
,-
Z 40 ,-'- .
U,I ,-
~ ,-

~
,-
30 ,-
£t: ,-
,-
,-
&... ,-
0 20 "P

~~
't- 10 ~
~
:~~~~~ ~~~~D

;';'
.~~ ,.," "
0 ---
2 4 6 8 10 2 4 6 10 12 14
DAYS FROM THE PICKING DATE
FIG. 4
Peach fruit (the first picking) keeping capacity in the house condition in response to water deficit applied during
different phenological phases of fruit development in 1987 (left) and in 1988 (right): Control (8-8). 1st
PWS treatment (0---0), 2nd PWS treatment (A-A), 3rd PWS treatment (6---6).1 +2 PWS
treatment (.-.) and 1+2+3 PWS treatment (0---0). Vertical bars represent L.S.D. at P = 0.05.

(Figure 4). Under home condition, fruits from
the trees of the 3rd PWS and 1+2+3 PWS
treatments rotted significantly less rapidly in
both years than did control fruits. At the end of
the fruit keeping trial (1(}-14d after picking),
about 4(}-50% of rotten fruits were recorded,
for the control treatment, but only 12-30% for
the first two treatments. Water stress applied
during the first two phases did not significantly
affect peach fruit keeping capacity. In 1987, no
significant differences were found between the
control treatment and the 1st PWS, 2nd PWS
and 1+2 PWS ones. In 1988, however, fewer
rotten fruits after picking were obtained on the
trees of the 1+ 2 PWS treatment as compared
with control fruits.
Fruit bud production
Water stress imposed on peach trees during
any stage of fruit development did not decrease
fruit bud formation (Table IV). On the con-
trary, the trees given treatments 2nd PWS, 1+2
PWS and 1+2+3 PWS produced significantly
more dense fruit buds than the control trees.
DISCUSSION
The final size of peach fruits depends
basically on the cell number and size in tb
mesocarp. Unlike shoot elongation and trun
or limb expansion growth, cell division in t~
mesocarp does not occur throughout fru
development, and cell enlargement is non-pn
gressive. Cell division occurs only during tl
first growth period, starting about 20 days aft,
full bloom and lasting about 40 days, whi
rapid cell expansion occurs during the la
growth phase (Jackson, 1968; Bollard, 197(
We have reported (Li et aI., 1989c and 1989
that, under conditions of intense water stre!
the growth rate of peach fruits was not at
affected during phase I, although a very lc
water potential of leaves and stomatal clost
were observed on these trees, while fruit exp~
sion was significantly limited by water den
during the final growth phase. This result
confirmed by the present study (Figure 3), SI
gesting that cell enlargement appears more $I
sitive to water stress than cell division (Hsi
1973; Begg and Turner, 1976). Moreover
period of water stress imposed on peach tr
during the first two phenological phases of fl
development favoured fruit growth a~
removal of the water stress status in the tr
(Figure 5). This favourable effect on f
 S.-H. LI, J.-G. HUGUET, P. G. SCHOCH and P. ORLANDO 549

CI
w 8 :J: Z
~
w
I
I
0::
0 ------ CONTROL
~ 0---0
..~
I-
~
6 --
-----
1 ST PWS
2 ST PWS
1+2 PWS
J:
I/)
w
0::
.....
4
I-
:5
0::
.....

.....
0
2
w
l-
e(
0::
>
:::! 0
~ Aug. Sept.
FIG. 5
Daily rate of fruit fresh weight increase during post-stress period in
1987. Fruit fresh weight was estimated according to fruit diameter:
y 2.99<°05'U.)(y: fruit fresh weight in g, x: fruit diameter in mm) which
=
is obtained in the laboratory by sampling fruits on the guard trees at
intervals. Vertical bars represent L.S.D. at P = 0.05.
growth during the post-stress period, also trees under treatments 1st PWS, 2nd PWS and
observed by Chalmers et ai. (1981) and Mitchell 1+ 2 PWS was never marked in 1987. As
and Chalmers (1982), is difficult to explain. regards shoot extension growth, neither after-
Since the rate of cell enlargement is dependent effect nor favourable action of water stress was
on its gross extensibility and turgidity status' evident during the post-stress period (Table II,
(Green, 1968; Hsiao, 1973) and the cell comparison between the control and 1st PWS).
turgidity status in those water-stressed fruits is Leaf growth is often reported to be quite
the same as in the control ones during the post- sensitive to water stress (Hsiao, 1973; Begg and
stress period, it is possible to suggest that cell Turner, 1976). Mild stress is enough to reduce
extensibility would be increased. This effect the development of leaf area. However, the
may be due to the violent changes of water expansion growth of leaves in peach trees is
status in the cells, from a good turgidity status only slightly sensitive to water stress (Li et af.,
.
to an important water deficit, or inversely 1989d). At the water deficit imposed in the
, during the period of water stress or at the present experi~ent, water stress had no effect
"

moment of water stress removal.
The response of shoot elongation and
increase in limb diameter to water stress is
clearly distinguished from that of fruit growth
as described above. On one hand, shoot
elongation and limb diameter increase. were
immediately inhibited whenever water supply
was restricted (Table II and Figure 2). On the
other hand, the after-effect of water stress on
limb growth was obvious (Figure 6). During the
post-stress period, growth recovery for the
on the leaf surface area (results non-reported).
It is evident from the present study that the
growth tolerance to drought varies with the dif-
ferent organs of peach trees (Li et ai., 1989d).
Based on the intensity of the growth inhibition
by water deficit; the sensitivity of organs to
water stress may be placed in the following
order of severity: limb diameter increase>
shoot elongation growth> fruit growth >
expansion of leaf area.
We had reported that a low level of water
 550 Soil water deficits and peach

E
=-.
300
8-- CONTROL

1&1 0 ---0 1 ST PWS

(/)
c(
1&1 ~2 ND PWS
It:
0 8-8 1'1-2 PWS
Z

It:
1&1
~
~
200
c(
C
CD 0 0

~ t, 1\ 1
::; I'
I ,", ~!1 I 1
~ I ~ "V,.
0100 i :' 0 I I , ,
1&1 ~ ~? "I I
I

~It: 'tJ
I" I
I
" I
I" I I
I I, I
0
i
0
V

1 ~ ~ i
~ I
1
I
I
PHASE III
~ i
I
PHASE II i
1
0
June July Aug. Sept.

FIG. 6
Daily rate of limb diameter increase during post-stress period in 1987.
Samples taken from dendrometer measurements on two trees in each
treatment before the onset of stem shrinkage. Each point represents the
mean of the growth of three days.

supply might partly prevent premature fruit
drop in peach trees compared with high levels
(Li and Huguet, 1989; Li et aI., 1989a). The
1987 results in the present study show that
water stress applied at any period also pre-
vented fruit drop from hand thinning after the
June physiological drop to maturity (Table IV).
In 1988, the premature fruit drop of the control
trees was slight, with a consequent absence of
;
significant differences in premature fruit drop
between the control trees and the water-
stressed ones. This phenomenon might be
explained by relatively dry soil conditions in the
control treatment plot. The micro-sprinklers
wetted only about 13% of the experimental
surface area. The remaining area was kept very
dry, particularly in 1988, as a result of the
covers protecting the experimental plot. This
explanation can be supported by a trial on the
same cultivar in the same orchard. Trees of the
same age under natural conditions without film
cover protection, also irrigated by micro-sprin-
klers according to soil water potential (I
siometermethod), had a39.5% prematuref
drop in 1988 (unpublished data).
The most pronounced effect of frequent i
gation and abundant water supply'
improved fruit size (Guelfat-Reich and B
Arie, 1980; Daniell, 1982; Panine and Meria
1985), the reduction in total soluble solid c
tent in the harvested fruits and in the f
storage capa<;ity (Guelfat-Reich et aI., IS
Guelfat-Reich and Ben-Arie, 1980; LOtte
al., 1985; Irving and Drost, 1987). Our resu
however, demonstrate that reduced W1
supply during the first two phases of f.
development did not affect final fruit Sizf
harvest (Table IV), which agrees with CI
mers etal. (1984), nor fruit quality (Table V:
Figure 4). The smaller size, higher level!
total soluble solids and longer storage dural
after picking were characteristic of the fr
from trees subjected to water stress at the
phase of fruit development.

,"
 S.-H. LI, J.-G. HUGUET, P. G. SCHOCH and P. ORLANDO 551

The density of flower bud production
increased on the trees of the treatments 2nd
PWS, 1+2 PWS and 1+2+3 PWS (Table IV).
The study of the rate of flower-induced buds
(ratio of flower buds to total buds) showed that
this high density of flower buds was due to the
stimulation of flower bud induction', in these
water-stressed trees (Table VI). A significant
correlation was found between the density of
flower bud (y) and the rate of flower-induced
bud (x):
y = -44.5 + 2.09x (r2 = 0.901, P<O.OOl).
Since full floral induction occurs from the *'end
of June to the first ten days of July for peach
trees in our region (Li et al., 1988 and 1989b),
we can suggest that flower bud formation can be
improved only by water deficit applied during
the critical period of flower bud induction.
Conclusion
There is considerable evidence that a period
of water deficit applied to peach trees during
particular stages of fruit development is some-
times beneficial. It is possible to control vigour
of peach trees without reducing fruit size and
yield and without affecting fruit quality by using
deficit irrigation during the first rapid fruit
growth and fruit pit hardening phases. Conse-
quently, water consumption and time of prun-
ing may be reduced. Since a short water stress
imposed during any phase can partly prevent
premature fruit drop, fruit production may
increase on water-stressed trees compared with
the normally irrigated ones. By improving fruit
size at harvest, water deficit during the first
phase is more useful than if applied during fruit
pit hardening or during the first two phases.
. Although water deficit during the final fruit
growth phase can improve fruit quality and
storage capacity, and reduce premature fruit
drop, the small fruits at harvest limit the appli-
cation of w~ter deficit during this period except
in peach cultivars with extra large fruit (e.g.
fruit of 170 or 200 g). For these, a slight reduc-
tion of fruit size, or even a fruit weight reduc-
tion of 20 or 30 g at harvest does not diminish
their market value. On the contrary, in some
countries such as France, fruits of moderate
size are sold more readily and at a good price.
We thank Mr C. Bussi (Engineer of INRA
Gotheron experimental orchard) for his con-
stant support during this work, Mr C. Billot
(Director of INRA Gotheron experimental
orchard) for the use of the facilities in the
experimental orchard, Miss L. Ron (INRA
Versailles, Documentation service) for her crit-
ical review of the manuscript and Mr M. Halna
duFretay and Miss F. Ressayre for their work in
, the plot.

REFERENCES
BEGG,J. E. and TURNER,N. C. (1976). Crop water deficits. Advances in Agronomy, 28, 161-217.
BOLLARD,E. G. (1970). The physiology and nutrition of developing fruits. In: The biochemistry of
fruits and their products (Hulme, A. C., Ed.). Academic Press, London and New York. 387-
425.
CHALMERS,D. J., MITCHELL,P. D. and HEEK,L. V. (1981). Control of peach tree growth and
productivity by regulated water supply, tree density and, summer pruning. Journal of
American Society for Horticultural Science, 106,307-12.
CHALMERS,D. J., MITCHELL,P. D. and JERIE,P. H. (1984). The physiology of growth control of
peach and pear trees using reduced irrigation. Acta Horticulturae, 146, 143--9.
DAGNELIE,P. (1975). Theorie et methodes statistiques. Vol. II. 2nd edition, Les Presses Agronomi-
ques de Gembloux, Gembloux.
DANIELL,J. W. (1982). Effect of trickle irrigation on the growth and yield of 'Loring' peach trees.
Journal of Horticultural Science, 57,393--9.
GREEN, P. B. (1968). Growth physics in Nitella: a method for continuous in vivo analysis of
extensibility based on a micro-manometer technique for turgor pressure. Plant Physiology,
43, 1169-84.
GRIGNAC,P. (1985). L'irrigation en grande culture. In: Agronomie mediterraneenne (Manuel de
L'Ecole Nationale Agronomique de Montpellier, ENSA Montpellier.
 552 Soil water deficits and peach

GUELFAT-REICH, S., ASSAF, R., BRAVDo, B. A. and LEVIN, I. (1974). The keeping quality of apples

in storage as affected by different irrigation regimes. Journal of Horticultural Science, 49,

217-25.
GUELFAT-REICH, S. and BEN-ARIE, R. (1980). Effect of irrigation on fruit quality at harvest and
during storage. Proceedings of XVth International Congress of Refrigeration, Vol. 111,423-7.
HILER, E. A., VAN BAVEL, C. H. M., HOSSAIN, M. M. and JORDAN, W. R. (1972). Sensitivity ofl
southern peas to plant water deficit at three growth stages. Agronomy Journal, 64, 60-4. j
HSIAO, T. C. (1973).Plant responses to water stress. Annual Review of Plant Physiology, 24,1
519-70. j
HUGUET, J.-G. (1985). Appreciation del'etat hydrique d'une plante it partir des variations micro-]
metriques de la dimension des fruits ou des tiges au cours de la journee. Agronomie, 5,733-
41. . :

IRVING,D. E. and DROST,J. H. (1987). Effects of water deficit on vegetative growth, fruit growth
and fruit quality in Cox's Orange Pippin apple. Journal of Horticultural Science, 62, 427-32.
JACKSON,D. I. (1968). Gibberellin aItd the growth of peach and apricot fruits. Australian Journal of
Biological Sciences, 21,209-15.
LANGLET,A. (1973). Effet de la secheresse sur la croissance et la production du sorgho grain.
Annales Agronomiques, 24, 307-38.
LI, S.-H., ATGER,P. and BUSSI,C. (1988). Flower physiological differentiation of peach as related
to twig type and bud position on the twig. Journal of Fruit Sciences,S, 1-5.
LI, S.-H. and HUGUET,J.-G. (1989). Production, qualite des fruits et croissance de pechers soumisa
differents regimes d'alimentation hydrique. Fruits, 44, 225-32.
LI, S.-H., HUGUET,J.-G. and BUSSI,C. (1989a). Irrigation scheduling in a mature peach orchard
using tensiometers and dendrometers. Irrigation and Drainage Systems, 3, 1-12.
LI, S.-H., BUSSI,c., HUGARD,J.and CLANET,H. (1989b). Critical period offlowerbud inductionin
peach trees associated with shoot length and bud position. Gartenbau Wissenschaft-Hor-
ticultural Science, 54, 49-53.
LI, S.-H., HUGUET, J.-G., SCHOCH,P. J. and BUSSI, C. (1989c). Reponse dejeunes pechers cultives
en pots it differents regimes d'alimentation hydrique. I: Consequence sur la transpiration, la
resistance stomatique, Ie potential hydrique foliaire, la photosynthese et les variations
micromorphometriques des tiges et fru\ts. Agronomie, 9, (in press).
LI, S.-H., HUGUET,J.-G., SCHOCH,P. J. and BUSSI,C. (1989d). Reponse de jeunes pechers cultives
en pots a differents regimes d'alimentation hydrique. II: Comportement de la croissance et
du developpement. Agronomie, 9, (in press).
LOTTER,J.DE V., BEUKES,D. J. and WEBER,H. W. (1985). Growth and quality of apples as affected
by different irrigation treatments. Journal of Horticultural Science, 60, 181-92.
MITCHAM,E. (1980). Thinned peaches are big peaches. Fruit Grower, N° March, 9.
MITCHELL,P. D. and CHALMERS, D. J. (1982). The effect of reduced water supply on peach tree
growth and yields. Journal of the American Society for Horticultural Science, 107, 853-6.
i PANINE,M. and MERlAux,S. (1985). Irrigation localisee du pecher en sol profond caillouteux de la
zone mediterraneenne. In: Les besoins en eau des cultures. .Conferences internationale Paris,
11-14 Sept. 1984. INRA CTIFL Press, Paris, 523-32.
ROMANI,R. J. and JENNING,W. G. (1971). Stone fruits. In: The biochemistry of fruits and their
products (Hulme, A. c., Ed.). Academic Press, London and New York, 411-36.
VIDAUD,J. and CLANET,H. (1978). L'eclaircissage des pechers et des nectaniers. INVUFLEC, 2nd
edition.
VIDAUD,J. and 41 other authors (1987). Le pecher: reference et techniques, CTIFL Press, Paris.
ZHENG,W. C. and 20 other authors (1979). Experimental methodology in fruit trees. Agricultural
Press, Peking.

(Accepted 20 June 1989)

."