Tree Physiology 16, 859--864

© 1996 Heron Publishing----Victoria, Canada

Water stress and crop load effects on fruit fresh and dry weights in
peach (Prunus persica)

M. E. BERMAN and T. M. DEJONG

Department of Pomology, University of California, Davis, CA 95616, USA

Downloaded from https://academic.oup.com/treephys/article/16/10/859/1668139 by guest on 06 June 2024

Received February 29, 1996

Summary Effects of water stress on fruit fresh and dry Crisosto et al. 1994), apple (Lotter et al. 1985) and Asian pear
weights were investigated in peach trees, Prunus persica (L.) (Caspari et al. 1994).
Batsch., with varying crop loads: light, moderate and heavy. In Horticultural studies of water stress effects on tree fruit
well-watered controls, tree water status was independent of growth have focused on fruit size, which is largely a measure
crop load. In trees receiving reduced irrigation, the degree of of fresh weight, but have not described the effects of water
water stress increased with increasing crop load. Water stress stress on dry weight accumulation. Research on tomato sug-
induced fruit fresh weight reductions at all crop loads. Fruit dry gests that water stress limits fleshy fruit water accumulation
weight was not reduced by water stress in trees having light to but does not affect carbon partitioning to the fruit (Ehret and
moderate crop loads, indicating that the degree of water stress Ho 1986, Mitchell et al. 1991). The effect of water stress on
imposed did not affect the dry weight sink strength of fruit. reproductive sink activity in tree crops has not been investi-
Water-stressed trees with heavy crop loads had significantly gated.
reduced fruit dry weights, which were likely due to carbohy- Fruit dry weight growth can be described in terms of the
drate source limitations resulting from large crop carbon de- realization of potential: i.e., the maximum growth possible
mands and water stress limitations on photosynthesis. given a non-limiting supply of resources (Wareing and Patrick
1975). Fruit growing at its potential rate is limited by its
Keywords: carbon partitioning, drought, fruit growth.
capacity for sink activity and, thus, is sink limited. When dry
weight accumulation is limited by insufficient carbohydrates,
growth is said to be source limited.
Introduction Grossman and DeJong (1995) have described the use of
The relationship between fruit growth and water stress is variable crop loads in peach to identify periods of source and
dynamic and depends on the developmental stage of the fruit, sink limitation to fruit dry weight growth. The maximum
the severity of water stress, and the component of growth being potential dry weight accumulation of individual fruit is deter-
considered. The growth of fleshy fruits such as apples, stone mined on trees where most fruit have been removed early in
fruits, and grapes consists of distinct developmental phases. In development. Growth of the remaining fruit is not limited by
peach (Prunus persica (L.) Batsch), rapid initial fruit growth is competition for carbohydrates and is thus sink limited. Fruit
followed by an intermediate phase of slow growth. This is dry weight growth on trees with heavier crop loads is then
followed by a period of very rapid fresh and dry weight compared to the potential maximum to identify source- and
increase that ends with maturity and ripening (Chalmers and sink-limited periods of growth. Dry weight accumulation
van den Ende 1975). During this final growth phase, which equal to the potential maximum indicates sink-limited growth.
consists of approximately one-third of the growth period, 65% Growth below potential is assumed to result from source limi-
of a fruit’s dry weight and 80% of a fruit’s fresh weight are tations.
accumulated (Chalmers and Wilson 1978). In peach, fruit on trees bearing a normal commercial crop
It has been observed that vegetative and reproductive growth load are usually source limited during the final stage of rapid
in trees are differentially sensitive to water stress. Additionally, growth (Pavel and DeJong 1993, Grossman and DeJong 1995).
reproductive growth is differentially sensitive to water stress at During this period, the sink demand of many rapidly growing
different times of the season. It has been reported that mild fruit is greater than assimilate supply (Grossman and DeJong
water stress applied during the intermediate developmental 1994).
period of slow fruit growth has no effect on crop yields but Water stress could potentially inhibit fruit dry weight growth
significantly reduces vegetative growth in peach (Mitchell and as a result of both sink and source limitations. Sink limitations
Chalmers 1982) and pear (Mitchell et al. 1984). However, the to fruit growth could occur if fruit cell expansive growth and
final period of very rapid fruit growth has been reported to be carbohydrate accumulation processes are sensitive to water
relatively sensitive to water stress in peach (Li et al. 1989, status. Source limitations could occur when water stress re-
860 BERMAN AND DEJONG

duces photosynthesis and restricts the supply of assimilates. light crop load (LC), minimum 10 cm between fruit, moderate
Moderate water stress will often reduce whole-plant carbon crop load (MC), minimum 5 cm between fruit (these trees had
gain, usually by inducing stomatal closure (Chaves 1991). standard commercial fruit spacing and crop load), heavy crop
The objective of this study was to investigate water stress load (HC), no fruit removed from tree. Trees were thinned
effects on peach fruit fresh and dry weight accumulation dur- during the first week of April, three weeks after full bloom.
ing the final stage of very rapid growth. These effects were Crop loads in the three thinning treatments were quite vari-
compared among trees with varying crop loads to determine: able. The LC treatment averaged 163 fruit per tree (S.D. =
(1) how much of the reported water stress-induced decrease in 58.97), the MC treatment averaged 265 fruit per tree (S.D. =
fruit size is attributable to fresh weight reduction versus dry 72.31) and the HC treatment averaged 561 fruit per tree (S.D.
weight reduction; (2) whether water stress reduces the poten- = 153.22).
tial dry weight accumulation of individual fruit on trees with
very light crop loads (i.e., does water stress induce sink limi- Water potential measurements

Downloaded from https://academic.oup.com/treephys/article/16/10/859/1668139 by guest on 06 June 2024

tations to fruit growth); and (3) whether source limitations to Water potential was measured with a Scholander pressure
fruit growth on trees with heavy crop loads are intensified by chamber (Soil Moisture Equipment Co., Santa Barbara, CA).
water stress. To measure whole-tree water status, water potentials were
measured at midday (from 1130 to 1230 h) on shaded leaves,
close to the main trunk, which had been bagged for at least one
Materials and methods hour before measurement. The leaf bags were plastic sheaths
covered with aluminum foil. This method of water potential
Plant material measurement eliminates the leaf-to-leaf variability encoun-
Before bloom, 96 trees from eight rows of 5-year-old ‘Elegant tered using exposed leaves by measuring a leaf in which water
Lady’ peach (P. persica) trees, on ‘Lovell’ rootstock, were potential has equilibrated with that of the main trunk
selected for uniformity, in a block at the UC Davis Wolfskill (McCutchan and Shackel 1992). Measurements were taken on
Experimental Orchard, Winters, California. The orchard was one leaf on each of four to six trees per thinning × irrigation
planted in a high density formation (5.5 × 2 m spacing) and treatment combination. Measurements were made at 5- to
trained to a Kearney perpendicular-V (DeJong et al. 1995). 9-day intervals during the last five weeks of the experiment.
Trees received standard commercial dormant pruning and 100
kg ha −1 N fertilization in the spring before the experiment. Gas exchange measurements
Because of technical difficulties, only one complete day of
Irrigation treatments photosynthetic measurements were collected (July 7), one
The experiment was set up as a split-plot with eight blocks week before harvest and one day after irrigation. Gas exchange
with irrigation being the main-plot factor. Eight pairs of adja- was measured three times during the day on mature, well-ex-
cent half-rows were selected as blocks. One half-row received posed outer canopy leaves of four trees from each of the six
the control treatment (CT) and the adjacent half-row received thinning and irrigation treatment combinations. Measurements
the water stress (WS) treatment. A furrow was ripped between were made with an ADC portable gas exchange system (ADC
rows to cut roots in the upper 40 cm of soil (the zone of most Ltd., Hoddesdon, U.K.).
root growth in this orchard) and to prevent surface water
movement between treatments. The trees were irrigated twice Fruit harvest
weekly by microjet sprinklers. Fruit were harvested on July 14, when the majority of the fruit
Reference evaporation (ET0) data for Winters, CA were in the LC treatment had reached maturity but before significant
obtained from the California Irrigation Management System drop occurred. Fruits in the other treatments were less devel-
(CIMIS). Before May 26, both treatments received 100% re- oped than those in the LC treatments, with the water-stressed
placement ET0 irrigation. From May 26 until the end of the HC trees having the least ripe fruits at time of harvest. All fruits
experiment on July 13, the CT trees received 120% replace- were removed from each tree and the total number of fruits
ment ET0 irrigation. Previous experiments (unpublished data) counted. The total crop fresh weight for each tree was meas-
indicated that significant time would elapse after imposition of ured and then a 10-fruit subsample from each tree was col-
reduced irrigation and observable effects on plant water status. lected. The sample was weighed, and dried at 65 °C in a forced
To impose water stress during the final four weeks of growth air draft oven, and dry weight recorded. From the fresh
in the WS treatment, irrigation was completely withheld in the weight/dry weight ratio of the subsample, the total crop dry
WS treatment for two weeks, from May 26 to June 9. From weight for each tree was calculated. The average fresh and dry
June 9 until the end of the experiment, WS trees received 25% weight per fruit was calculated by dividing total crop fresh
replacement ET0 irrigation. weight and total crop dry weight by total fruit number.

Thinning treatments Data analysis

Within each irrigation plot, the CT and WS rows were divided All mean and standard error determinations, means separation
into three sub-plots, consisting of pairs of adjacent trees. Sub- tests, and ANOVAs were done using SAS statistical software
plots were randomly assigned one of three thinning treatments: (SAS Institute, Cary, NC). Photosynthetic rates and water
 WATER STRESS AND PEACH FRUIT GROWTH 861

potentials were compared for each irrigation × thinning treat- 4A): i.e., by 23, 26 and 37% in the LC, MC and HC thinning
ment using Tukey’s means separation test at a significance treatments, respectively. In both irrigation treatments, adjusted
level of P < 0.05. mean fruit fresh weight decreased with increasing crop load.
The yield data were analyzed as a split-plot design. Data
from pairs of trees in each sub-plot were averaged to yield a
single replication value for each sub-plot. Significant irrigation
(main plot) and thinning (sub-plot) effects were observed for
both fresh and dry crop weights. To analyze the data in greater
detail, the experiment was stratified by thinning treatment and
each thinning treatment analyzed separately.
Because of varying tree sizes and degrees of fruit set, there
was considerable variability in tree-to-tree crop load within

Downloaded from https://academic.oup.com/treephys/article/16/10/859/1668139 by guest on 06 June 2024

each thinning treatment. To reduce the variance associated
with variable crop loads, fresh and dry fruit weight means for
each thinning treatment were adjusted using the difference
between mean thinning treatment crop load (163, 265, and 561
fruit per tree for the LC, MC and HC treatments, respectively)
and actual crop load as a covariate. This type of analysis gave
an adjusted mean and variance which would have been ob-
served if all trees within each thinning treatment had equal
crop loads (Steel and Torrie, 1980). Each thinning treatment
was analyzed by ANOVA covariate analysis to determine the
significance of irrigation effects and to compute adjusted
means (Freund et al. 1986). Irrigation effects were considered
significant at P < 0.05. Figure 1. Seasonal pattern of midday stem water potential for each
irrigation and thinning treatment combination. The irrigation treat-
ments are control treatment (CT) and water stress (WS) treatment. The
Results thinning treatments are light crop (LC) with a mean crop load of 163
fruit per tree, moderate crop (MC) with a mean crop load of 265 fruit
per tree, and heavy crop (HC) with a mean of 561 fruit per tree. Each
Water stress point represents the mean of four to six measurements. Error bars
In early June, midday stem water potential differences between represent standard errors.
the two irrigation treatments were observed (Figure 1). The
average stem water potential during the last five weeks of the
season was differentially affected by crop load in the two
irrigation treatments (Figure 2). Within the CT treatment, stem
water potential was independent of load treatment. Within the
WS treatment, water potential became more negative with
increasing crop load and was significantly lower in HC trees
than in LC trees.

Gas exchange
Carbon dioxide assimilation rates were highest early in the day
and declined in the afternoon (Figure 3). Leaf assimilation
rates in the WS treatment equaled those in the CT treatment
early in the day. In the afternoon, the difference between the
two irrigation treatments was greatest, with WS trees having
significantly lower late afternoon leaf assimilation rates than
CT trees in the MC and HC treatments. Reduced leaf assimi-
lation rates were highly correlated with low water potentials
and leaf conductances (data not shown). Over the 7.5 hour
measurement period, the total estimated leaf assimilation inte-
gral in the WS treatment was reduced by 21% in the MC trees
and 27% in the HC trees.
Figure 2. Mean bagged-leaf water potential for each irrigation and
Individual fruit and total crop weight thinning treatment combination during the last five weeks of fruit
development. Error bars represent standard errors. Means not labeled
Adjusted individual mean fruit fresh weight was significantly with a common letter are significantly different from one another
lower in the WS treatment than in the CT treatment (Figure (Tukey’s Means Separation Test, P < 0.05).
862 BERMAN AND DEJONG

Downloaded from https://academic.oup.com/treephys/article/16/10/859/1668139 by guest on 06 June 2024

Figure 3. Carbon dioxide assimilation rates on July 7. Each point
represents the mean of four leaf measurements, each on a separate tree. Figure 4. Adjusted mean individual fruit fresh and dry weights. Error
The same trees were measured at three times throughout the day. Error bars represent adjusted standard errors. Asterisks represent significant
bars represent standard errors. Asterisks represent significant irriga- ANOVA effects for irrigation treatment at P < 0.05. (A) Individual
tion treatment differences at each time point (Tukey’s Means Separa- fruit fresh weight, (B) individual fruit dry weight.
tion Test, P < 0.05). (A) LC treatment, (B) MC treatment, (C) HC
treatment.

Irrigation treatment had no significant effect on adjusted mers et al. 1983). DeJong (1986) compared gas exchange
individual mean fruit dry weight in the LC and MC treatments parameters of the leaves of fruiting and non-fruiting peach
(Figure 4B). In the HC treatment, WS irrigation significantly trees and observed that fruit induced 30% greater stomatal
reduced adjusted mean dry fruit weight by over 18%. conductance and 11--15% increased photosynthetic rates. In-
Adjusted mean total crop fresh weight was significantly creased water use due to crop-induced transpirational in-
greater in CT trees than WS trees for all thinning treatments creases may account for the water potential differences
(Figure 5A). Adjusted mean dry crop weight was not signifi- observed in this study. In the CT trees, increased fruit loads
cantly different between irrigation treatments in the LC and apparently did not affect water status because sufficient soil
MC thinning treatments, but was significantly reduced by 17% water was present at all times.
in the water-stressed HC trees (Figure 5B). The increased water stress associated with larger crop loads
could also be the result of reduced root growth. Williamson and
Coston (1989) observed that even light crop loads significantly
Discussion reduced root growth in peach during the period of maximal
fruit growth. Because soils in the WS irrigation treatment
Water stress dried, root growth to exploit increased soil volume may have
Within the WS irrigation treatment, the severity of water stress been inhibited by insufficient carbon supply in heavily crop-
was dependent on crop load, whereas in the CT treatment, ping trees.
water status was independent of load (Figure 2). It has been
Fruit fresh weight
observed in peach that cropping trees are often more water
stressed than defruited trees (Chalmers and Wilson 1978). Water stress caused a significant decrease in fruit fresh weight
During the final growth phase, when assimilate demand by (Figure 4A). This is in agreement with horticultural studies
fruit is at a maximum, leaf transpiration rates are reported to where peach fruit size, largely a measure of fresh weight, was
be higher on cropping trees than on non-cropping trees (Chal- reduced by water stress (Li et al. 1989, Crisosto et al. 1994).
 WATER STRESS AND PEACH FRUIT GROWTH 863

trees (Figure 2) or from the significant water stress effects on

the dry weight accumulation of these fruit. Fresh weight
growth depends on the accumulation of large quantities of
osmotically active solutes and massive cell expansive growth
(Zucconi, 1986). These processes require carbohydrates, and
apparently restricted carbohydrate supply of the water-stressed
HC trees would further decrease fruit ability to accumulate
water.

Crop and fruit dry weight

Within the LC load treatment, irrigation treatment had no

Downloaded from https://academic.oup.com/treephys/article/16/10/859/1668139 by guest on 06 June 2024

effect on fruit dry weight (Figure 4B). The LC trees were
heavily thinned and fruit growth in this treatment should have
been sink-limited (Grossman and DeJong 1995). The water
stress experienced by the LC trees was not great enough to
affect the sink strength of fruit. In the water-stressed MC
treatment, the water potentials were more negative, yet no
significant reduction in dry weight was observed relative to
well-watered MC controls. This indicates that fruit sink
strength was not reduced in trees with water potentials averag-
ing 0.45 MPa below those of fully watered controls. This is in
agreement with work with tomato, in which water stress de-
creases fruit fresh weight without decreasing fruit sink strength
(Ehret and Ho 1986, Mitchell et al. 1991). Therefore, peach
fruit resemble tomato fruit in that fresh and dry weight accu-
mulation are not necessarily correlated (Grange and Andrews,
1995).
Figure 5. Adjusted mean crop fresh and dry weights. Error bars Fruit dry weight was significantly reduced by water stress in
represent adjusted standard errors. Asterisks represent significant the HC treatment. It is unlikely that water stress induced sink
ANOVA effects for irrigation treatment at P < 0.05. (A) Crop fresh limitations to fruit growth in this treatment, unless a threshold
weight, (B) crop dry weight. water potential that reduces fruit sink strength exists between
−1.21 MPa (that of the MC trees where no sink limitations
were observed) and −1.33 MPa (that of the HC trees). There-
Reduced fruit fresh weight as a result of water stress is ex- fore we assume that total sink demand in the HC treatment was
pected. Water potential is highly correlated with plant relative not affected by irrigation.
water content (RWC) (Koide et al. 1991), and the WS trees, Assuming equal carbon demand in HC trees from both
having lower water potentials than the CT trees, would be irrigation treatments, the significantly reduced total crop dry
expected to have reduced whole-plant RWC. However, the weight in WS trees (Figure 5B) represents a source limitation
specific mechanism of fruit fresh weight reduction is unclear. to dry weight accumulation. The total crop dry weight in the
Fleshy fruit water relations, especially near fruit maturity, CT-HC trees can be considered the maximum potential carbo-
are complex. In grape (Greenspan et al. 1994) and tomato (Ho hydrate supply and the 17% difference between CT and WS
et al. 1987), the maturing fruit may be partially isolated from dry crop weights represents the reduction in carbohydrate
the apoplast of the plant, and most water is supplied through supply caused by water stress.
the phloem. However, apoplastic water status strongly influ- This is in agreement with the leaf CO2 assimilation data
ences water flow in the phloem. The water potential gradient (Figure 3). Water stress in the WS-HC trees led to stomatal
between the fruit and the plant controls phloem-driven expan- closure and significantly reduced assimilation rates, relative to
sive growth of tomato fruit (Johnson et al. 1992). In Phaseolus, CT controls. Steinberg et al. (1990) observed that water poten-
translocation flow from the leaves to the roots was observed to tials 0.6 MPa below those of fully watered controls caused
be proportional to the source−sink water potential gradient severe reductions in leaf conductance and biomass production
(Lang and Thorpe 1986). In this study, the reduced water in young peach trees. In the HC trees, water-stress-induced
potential of the WS treatment trees may have reduced the reductions in assimilation, combined with high demand for
driving force for water flow into the fruit. carbohydrates appeared to create a source limitation to fruit
The fresh weight reductions were 23 and 26% in the LC and dry weight growth. Water-stress-induced reductions in leaf
MC treatments, respectively, and over 37% in the HC treat- assimilation rates were also observed in the MC treatment;
ments. This greater degree of fresh weight reduction may have however, the crop load, and thus the total sink demand of the
resulted from the more severe water stress observed in the HC fruit in this treatment, was less than that of the HC trees.
864 BERMAN AND DEJONG

Adequate carbon was apparently available to support a moder- Grossman, Y.L. and T.M. DeJong. 1994. PEACH: A simulation model
ate crop load, even with reduced photosynthesis. of reproductive and vegetative growth in peach trees. Tree Physiol.
A limitation of this study is that fruit within the various 14:329--345.
treatments were at different developmental stages when har- Grossman, Y.L. and T.M. DeJong. 1995. Maximum fruit growth po-
tential and seasonal patterns of resource dynamics during the
vested. The LC fruit in both irrigation treatments were at full
growth of individual fruit. Ann. Bot. 75:553--560.
maturity when the fruit were harvested whereas fruit develop- Ho, L.C., R.I. Grange and A.J. Picken. 1987. An analysis of the
ment in the other treatments appeared to be 3--6 days behind. accumulation of water and dry matter in tomato fruit. Plant Cell
Thus the study measured the effects of water stress and crop Environ. 10:157--162
load on fresh and dry weight accumulation rate, rather than Johnson, R.W., M.A. Dixon and D.R. Lee. 1992. Water relations of the
water stress effects on final fruit size. Fruit growth rate tends tomato during fruit growth. Plant Cell Environ. 15:947--953.
to be highest just before fruit maturity is reached (Grossman Koide, R.T., R.H. Robichaux, S.R. Morse and C.M. Smith. 1991. Plant
and DeJong 1995). If development stage had been a factor in water status, hydraulic resistance and capacitance. In Plant Physi-

Downloaded from https://academic.oup.com/treephys/article/16/10/859/1668139 by guest on 06 June 2024

this experiment, one would have expected greater treatment ological Ecology. Eds. R.W. Pearcy, J. Ehleringer, H.A. Moony and
effects on fruits growing at the highest rates. The greatest P.W. Rundel. Chapman and Hall, London, pp 161--183.
treatment effects were actually on the HC trees, which had the Lang, A. and M.R. Thorpe. 1986. Water potential, translocation and
assimilate partitioning. J. Exp. Bot. 37:495--503.
most retarded development.
Li, S.H., J.G. Huguet, P.G. Schoch and P. Orlando. 1989. Response of
peach tree growth and cropping to soil water deficit at various
References phenological stages of fruit development. J. Hort. Sci. 64:541--552.
Lotter, J.D.V., D.J. Beukes and H.W. Weber. 1985. Growth and quality
Caspari, H.W., M. Hossein and D.J. Chalmers. 1994. Water use, of apples as affected by different irrigation treatments. J. Hort. Sci.
growth, and fruit yield of ‘Hosui’ Asian pears under deficit irriga- 60:181--192.
tion. J. Am. Soc. Hort. Sci. 119:383--388. McCutchan, H. and K.A. Shackel. 1992. Stem-water potential as a
Chalmers, D.J. and B. van den Ende. 1975. A reappraisal of the growth sensitive indicator of water stress in prune trees (Prunus domestica
and development of peach fruit. Aust. J. Plant Physiol. 2:623--634. L. cv. French). J. Am. Soc. Hort. Sci. 117:607--611.
Chalmers, D.J. and I.B. Wilson. 1978. Productivity of peach trees: Mitchell, J.P., C. Shennan and S.R. Grattan. 1991. Developmental
Tree growth and water stress in relation to fruit growth and assimi- changes in tomato fruit composition in response to water deficit and
late demand. Ann. Bot. 42:285--294. salinity. Physiol. Plant. 83:177--185.
Chalmers, D.J., K.A. Olsson and T.R. Jones. 1983. Water relations of Mitchell, P.D. and D.J. Chalmers. 1982. The effect of reduced water
peach trees and orchards. In Water Deficits and Plant Growth, Vol. supply on peach tree growth and yields. J. Am. Soc. Hort. Sci. 107:
VII. Ed. D.J. Chalmers. Academic Press, NY, pp 197--232. 853--856.
Chaves, M.M. 1991. Effects of water deficits on carbon assimilation. Mitchell, P.D., P.H. Jerie, and D.J. Chalmers. 1984. The effects of
J. Exp. Bot. 42:1--16. regulated deficit irrigation on pear tree growth, flowering, fruit
Crisosto, C.H., R.S. Johnson, J.G. Luza and G.M. Crisosto. 1994. growth, and yield. J. Am. Soc. Hort. Sci. 109:604--606.
Irrigation regimes affect fruit soluble solids concentration and rate Pavel E.W. and T.M. DeJong. 1993. Source- and sink-limited growth
of water loss of ‘O’Henry’ peaches. HortScience 29:1169--1171. periods of developing peach fruits indicated by relative growth rate
DeJong, T.M. 1986. Fruit effects on photosynthesis in Prunus persica. analysis. J. Am. Soc. Hort. Sci. 188:820--824.
Physiol. Plant. 66:149--153. Steel, G.D. and J.H. Torrie, 1980. Principles and procedures of statis-
DeJong, T.M, K.R. Day, J.F. Doyle and R.S. Johnson. 1995. The tics. McGraw-Hill, USA, pp 401--437.
Kearney Agricultural Center perpendicular ‘‘V’’ (KAC-V) orchard Steinberg, S.L., J.C. Miller and M.J. McFarland. 1990. Dry matter
system for peaches and nectarines. Hort. Tech. 4:362--367. partitioning and vegetative growth of young peach trees under water
Ehret, D.L. and L.C. Ho. 1986. Effects of osmotic potential in nutrient stress. Aust. J. Plant Physiol. 17:23--36.
solution on diurnal growth of tomato fruit. J. Exp. Bot. 37:1294-- Wareing, P.F. and J.W. Patrick. 1975. Source-sink relations and the
1302. partitioning of assimilates in the plant. In Photosynthesis and Pro-
Freund, R.J., R.C. Littell and P.C. Spector. 1986. SAS System for ductivity in Different Environments. Ed. J.P. Cooper. Cambridge
linear models. SAS Institute, Cary, N.C., pp 85--141. University Press, pp 481--499.
Grange, R.I. and J. Andrews. 1995. Respiration and growth of tomato Williamson, J.G. and D.C. Coston. 1989. The relationship among root
fruit. Plant Cell Environ. 18:925--930. growth, shoot growth and fruit growth of peach. J. Am. Soc. Hort.
Greenspan, M.D., K.A. Shackel and M.A. Matthews. 1994. Develop- Sci. 114:180--183.
mental changes in the diurnal water budget of the grape berry Zucconi, F. 1986. Peach. In Handbook of Fruit Set and Development.
exposed to water deficits. Plant Cell Environ. 17:811--820. Ed. S.P. Monselise. CRC Press, Boca Raton, FL, pp 303--321.