Summary Net photosynthetic rates (A) of leaves on 11-year-old, field-grown apple trees (Malus domestica Borkh. cv. Gold-en Delicious) were measured after removal of fruits at four different stages of development. Defruiting decreased A by 21, 42, 27 and 7% when fruits were growing at 311, 293, 229 and 113 mgDW day−1, respectively. Photosynthesis was inhibited more in the afternoon than in the morning, but it was not affected during the first 8 h after fruit removal. Inhibition of A was positively correlated with crop sink strength, but it was not correlated with fruit relative growth rate or crop load. Defruit-ing decreased A at saturating irradiances (PPFD > 1000 µmol m−2 s−1), but did not modify the apparent quantum yield of single leaves. These results suggest that the overall effect of defruiting on carbon fixation is negligible in dense canopies, but it may be significant in sparse canopies and in single shoots. Keywords: apparent quantum yield, fruit growth rate, Malus domestica, sink strength.
Introduction
Fruit removal results in a decreased leaf net photosynthetic rate (A) in several tree fruit species, including apple (Malus domes-tica Borkh.) (Hansen 1970, Monselise and Lenz 1980, Schaf-fer et al. 1987, Gucci et al. 1991). A 20--40% decline in A appears as early as 16 h after fruit removal in leaves of field-grown plum (Prunus domestica L.) trees and persists for 2 to 3 weeks; however, the decrease is less evident or absent when vegetative sinks are actively developing (Gucci et al. 1991). In other studies, fruiting trees show either a 15--20% increase in A only during periods of high fruit growth (Fujii and Kennedy 1985, DeJong 1986) or no effect (Proctor et al. 1976, Roper et al. 1988, Schechter et al. 1991). To explain these discrepancies, it has been hypothesized that total carbohydrate demand by the tree at the time of fruit detachment may be a critical factor in determining the photosynthetic response of leaves (Ferree and Palmer 1982, Gucci and Flore 1992). Thus, both fruit growth rate and crop size at the time of fruit removal could affect the
magnitude of depression of leaf photosynthesis induced by defruiting.
Fruit removal has also been shown to decrease stomatal conductance and carboxylation efficiency, to decrease specific leaf area, and to alter carbohydrate partitioning in the leaf mesophyll (Proctor et al. 1976, Monselise and Lenz 1980, Schaffer et al. 1987, Gucci et al. 1991). These changes may be interpreted as the result of interrupted utilization of carbon assimilates by the fruit, which eventually causes a decrease in A through a feedback mechanism (Guinn and Mauney 1980). An impairment in the carbon reduction cycle would also result in lower light-saturated A but unaltered quantum yield (Ehler-inger and Bjorkman 1977). Although light-saturated A gener-ally decreases following fruit removal (Gucci et al. 1991), there are no reports of effects of defruiting on quantum yield and A at nonsaturating irradiances. Because only a fraction of the total leaf area of field-grown trees is fully exposed to sunlight at any given time, a change in apparent quantum yield following defruiting may indicate whether the depression of photosynthesis observed in single leaves also occurs over the whole canopy.
The objective of this study was to determine whether a relationship exists between changes in leaf A following de-fruiting and crop sink strength, the latter defined as the product of fruit relative growth rate and crop load. We also investigated the response of CO2 assimilation rate to incident photon flux density in field-grown ‘‘Golden Delicious’’ apple trees after fruit removal.
Materials and methods
Plant material
In spring 1990, 16 11-year-old apple trees (cv. Golden Deli-cious on M27 rootstock), growing in an orchard on a fertile, clay-loam soil at Cadriano (BO), Italy (latitude 45° N, altitude 80 m a.s.l.) were selected for uniformity in leaf/fruit ratio and crop load. Eight additional trees from the same orchard were
The effect of defruiting at different stages of fruit development on leaf
photosynthesis of ‘‘Golden Delicious’’ apple
R. GUCCI
1, L. CORELLI GRAPPADELLI
2, S. TUSTIN
2,3and G. RAVAGLIA
21 Dipartimento di Coltivazione e Difesa delle Specie Legnose, Sez. Coltivazioni Arboree, University of Pisa, Pisa, Italy
2 Dipartimento di Colture Arboree, University of Bologna, Bologna, Italy
3 Present address: The Horticulture and Food Research Institute of New Zealand, Havelock North Research Centre, Private Bag 1401, Havelock
North, New Zealand
Received March 4, 1994
also selected in spring 1991. The trees were spaced 0.75 × 1.5 m apart in a N--S oriented, six-row bed and trained to a full-field minibush system (Preston 1978). The trees were periodically sprayed with a commercial pesticide and were sprinkle-irrigated to supplement natural rainfall during the growing season. During the gas exchange measurements, irri-gation was only used on June 28, 1990, because it rained abundantly the week before fruit removal on the other dates. The number of fruits was adjusted to 60--70 and 45--70 per tree by hand-thinning on May 20, 1990 and June 12, 1991, respec-tively. Thus, crop load per tree depended solely on changes in fruit growth rate or fruit size and not on number of fruits. In 1990, full bloom occurred on April 2, petal fall occurred on April 9, termination of shoot elongation occurred 6--8 weeks after full bloom, and the crop was commercially harvested on September 14. In 1991, full bloom and commercial harvest occurred on April 1 and September 13, respectively.
In 1990, three trees were completely defruited on four dates beginning with the period of maximum fruit growth rate and ending with the commercial harvest (see Figure 1). The re-maining four trees served as controls. In 1991, four trees were defruited on June 24 and four were left intact to serve as controls. In both years, fruits were removed by hand between 0830 and 0945 h, except on September 10, 1990, when they were removed at 1115 h.
To determine the pattern of fruit growth, samples of 10 to 15 fruits were randomly selected and harvested weekly from May 31 to September 14, 1990, from trees adjacent to those chosen for gas exchange measurements, and their equatorial diameters and fresh weights recorded. A similar sampling procedure was followed in 1991. Dry weight/fresh weight ratios were deter-mined from a section of each fruit after oven drying to a constant weight, and then dry weights were calculated for individual fruits. Absolute growth rate of fruits was calculated as the first derivative of the sigmoidal curves fitted to the fresh and dry weight functions of the general type:
y=a exp[−b exp(−cx)],
where a, b and c are coefficients, and y and x are the dependent and independent variables, respectively.
Gas exchange measurements
Gas exchange parameters, leaf temperature and photosynthetic photon flux density (PPFD) were measured in the field with an ADC LCA-2 portable open gas exchange unit, equipped with a Parkinson broadleaf chamber and a mass flow air supply unit (Analytical Development Co., Hoddesdon, U.K.), operated at the following conditions unless otherwise stated: PPFD > 1000
µmol m−2 s−1, an ambient CO2 concentration of 325--355 µl l−1, and a flow rate of 0.4 l min−1.
Four fully expanded leaves were chosen on the shoot emerg-ing at the axil of the youngest leaf of the spur whorl (bourse shooot) of two to four fruiting spurs in the upper part of the canopy of each tree at the beginning of the experiment. Se-lected leaves, which were located within 15 cm of one or more fruits, remained exposed to full sunlight during most of the day
and hence, it was possible to measure diurnal changes in light-saturated net photosynthesis. To reduce sample variabil-ity, the same leaves were measured on each occasion of fruit removal, between 0900 and 1130 h and between 1230 and 1530 h.
A total of 16 response curves of net photosynthesis to incident PPFD (A/Q) were measured in the field on 3--4 leaves per tree on August 9, 1990, September 10 and 12, 1990, and June 26, 1991. Measurements usually began at 1030 h, when stomata appeared to be fully open. Using natural sunlight as the light source, PPFDs ranging in value from about 1800 to 10 µmol m−2 s−1 were obtained in six to eight steps by shading the leaf chamber with various layers of black neutral density shade cloth or translucent paper. Shading materials were checked for neutrality of absorbance in the 400--700 nm wave-length range with an LI188 spectroradiometer (Li-Cor, Lin-coln, NE, USA). Checks on the variation of A to the highest and lowest PPFDs were made by returning to saturating irradi-ance after a complete A/Q series was measured. The rate of photosynthesis was usually within 10% of the initial value. To avoid overheating of the leaf chamber, which was not tempera-ture controlled, the cuvette was removed from the leaf between measurements, the leaf acclimated for 60 s to the next irradi-ance, and then the cuvette was clamped onto the leaf again. The data were recorded once the differential CO2 had equilibrated. Temperature changes were within 2.5 °C during each A/Q response curve, whereas stomatal conductance decreased less than 15% from the highest to lowest PPFD. Leaves of adjacent defruited and control trees were alternately measured to mini-mize diurnal effects on treatment response during A/Q meas-urements. It took about 60 min to obtain a pair of A/Q curves. Apparent quantum yield was estimated as the slope of the regression equation for each set of data in the linear range of response (0--200 µmol m−2 s−1).
Statistical analysis
The experimental design was a randomized complete block, with four blocks of four trees each, arranged according to location in the orchard. At four dates during the 1990 growing season, three trees were randomly selected and defruited, and their photosynthetic performance compared with that of con-trol trees by two-way ANOVA within each date of measure-ment. Gas exchange parameters of four leaves per tree were measured and averaged. Slopes of linear regression equations were tested for significance by Student’s t-test (Steel and Torrie 1980).
Results
growth rate was attained (Table 1).
Effects of fruit removal on leaf A depended on the time of treatment (Figure 2). No effect of fruit removal was evident during the first 8 h after defruiting, except on July 17 when significant differences in leaf A appeared 3 h after defruiting. However, no differences were detected at 2, 4.5 and 7 h after defruiting on July 17 (Figure 2b). Differences in leaf A be-tween fruiting and defruited trees usually began to appear within 24 h of defruiting and became statistically significant between 24 and 48 h after defruiting (Figure 2 and Table 2). Assimilation rates of control trees remained more or less constant throughout the growing season (Figure 2), and no effect of leaf age on A was detected. Assimilation was usually at a maximum at 1000--1100 h and tended to decline in the
afternoon. The inhibition of A in response to defruiting was more marked in the afternoon than in the morning (Table 2). The decrease in A was greatest when fruit was removed on July 17 and least when fruit was removed on September 10 (Ta-ble 1). Crop sink strength was about 18 gDW day−1 tree−1 over the first three dates of defruiting, but decreased to 6 gDW day−1 tree−1 by September 10 (Table 1).
Analysis of A/Q curves showed that fruit removal depressed A up to 43% at saturating irradiance and by 40 to 0% between the light saturation point and 400 µmol m−2 s−1 (Figure 3). In general, apparent quantum yield of control trees was only marginally higher (5--12%) than that of defruited trees (Ta-ble 3), and the difference was not significantly different (P > 0.05). Assimilation at PPFD values between 250 and 450 µmol m−2 s−1 (avoiding sun flecks) was about 35--50% of that at saturating light intensity. No differences in A between de-fruited and fruiting trees were found at irradiances below 400
µmol m−2 s−1 (data not shown).
Discussion
Defruiting decreased single leaf A differently depending on the stage of development of the fruit when removed. Maximum inhibition (42%) was similar to values reported for apple by Monselise and Lenz (1980), but intermediate between the values reported by Hansen (1970) and Fujii and Kennedy (1985). However, a comparison of the homologous periods (based on full bloom date, end of leaf expansion and pattern of fruit growth) of the 1981 growing season at Pullman, Washing-ton, USA (Fujii and Kennedy 1985) and the 1990 season at Bologna, Italy, indicated that the June, July and August dates of fruit removal in our study (A inhibited more than 20%) corresponded to the period when 20% differences in A were measured between fruiting and nonfruiting trees in the study by Fujii and Kennedy. Also, fruit removal at the time of commercial harvest (7% inhibition) in our study coincided with a period of no differences in A in the Washington study. Thus, despite different experimental protocols and climatic conditions, the results from these two studies are similar.
Air temperature, quantum flux density and distribution of rainfall were similar during the four periods of gas exchange measurements in 1990, and leaf age did not affect the photo-synthetic potential of leaves from June to September. There-Figure 1. Curves of equatorial diameter, cumulative fresh weight
(FW) and fresh weight growth rate (AGR) of ‘‘Golden Delicious’’ apple fruits from May 31 to September 14, 1990. Arrows pointed downward indicate dates of fruit removal and numbers the relative sequence. Arrows pointed upward indicate end of shoot elongation (E) and date of commercial harvest (H). Horizontal bars at the bottom of figure indicate periods during which gas exchange parameters were measured. Data points and vertical bars are means ± SD of 10--15 replicates (bars not shown if smaller than symbols). The diameter curve was fitted by the equation: diameter = −655.9 exp(0.017 × day) + 85.7, R2 = 0.996. The growth rate curve was calculated as the first derivative of the cumulative fresh weight equation: FW = 207.1 exp[−137.9 exp(−0.030 × day)], R2 = 0.99.
Table 1. Crop load, fruit growth rates and crop sink strength at four stages of fruit development of ‘‘Golden Delicious’’ apple in 1990. Crop sink strength was calculated as the product of fruit relative growth rate and crop load. Inhibition of photosynthesis is the mean of values calculated from all measurements on each date of fruit removal.
Date of Days after Crop load1 Fruit absolute Fruit relative Crop sink Inhibition of fruit removal full bloom (gDW tree−1) growth rate2 growth rate strength photosynthesis
(mgDW day−1) (mgDW g−1 day−1) (gDW day−1 tree−1) (%)
June 28 87 541 311 33 17.8 21
July 17 106 858 293 21 17.8 42
August 6 126 1310 229 14 18.2 27
September 10 161 1450 113 4 5.8 7
fore, although in field experiments it is impossible to separate unequivocally the effect of environmental conditions on pho-tosynthesis from factors endogenous to the tree, we suggest that the extent of inhibition of A in response to defruiting differed between the September measurements and those taken earlier because of differences in crop sink strength (Table 1). Inhibition of A increased progressively with time after de-fruiting. The absence of an immediate response may explain why Proctor et al. (1976), who measured A for only 0.5 h after fruit removal, did not detect a change in A. Defruiting treat-ments did not affect A until more than 24 h after fruit removal, and the effect was greater during the afternoon than during the morning (Figure 2 and Table 2). This time course resembled that observed in plum, where a decline in A was not detected until 16 h after defruiting (Gucci et al. 1991). In our study, limitations on tree growth imposed by the M27 rootstock likely played a critical role in the photosynthetic response to fruit removal. Differences in A may not have developed in apple trees grafted on more vigorous rootstocks, where alternative sinks could have been present to absorb the carbon surplus after fruit removal. We measured photosynthetic rates that were slightly lower than the mean value of 16.8 µmol m−2 s−1 reported for apple on M27 rootstock (Schechter et al. 1991).
Apparent quantum yields of control and defruited trees were
lower than those reported previously for apple (Schechter et al. 1991). High temperature (> 30 °C) and vapor pressure deficit in the cuvette during A/Q measurements may have depressed these values. Apparent quantum yield has been shown to de-crease with temperatures above 14 °C in C3 plants (Ehleringer and Bjorkman 1977), and previous studies on apple were conducted at lower temperatures (25--29 °C) (Schechter et al. 1991) than ours. The finding that quantum yield was unaf-fected by fruit removal, whereas A was markedly inhibited at saturating light intensity, has important implications. First, it is compatible with the hypothesis of feedback regulation of photosynthesis by carbon products because, at PPFD < 200
µmol m−2 s−1, utilization of excitation energy is matched by a similar rate of carbon metabolism, whereas at saturating irra-diances, carbon metabolism lags the utilization of excitation energy as long as light harvesting continues unaffected (Ehler-inger and Bjorkman 1977). Second, the pattern of light re-sponse curves also suggests that the 21--42% average inhibition of A, which we observed on leaves fully exposed to light after fruit removal during active fruit growth, becomes much smaller and may even disappear at the whole canopy level. Light flux density varies spatially and diurnally within the tree canopy (Barden 1974), and for most training systems adopted for apple trees, a large portion of the foliage is exposed Figure 2. Changes in leaf net photosynthetic rate (A) of field-grown ‘‘Golden Delicious’’ apple trees after fruit removal at four stages of fruit development in 1990. Arrows indicate the time of fruit removal and relative numbers correspond to the stage of fruit growth in Figure 1. Data points are means of three trees per treatment (SE bars not shown if smaller than symbol). Asterisks refer to means significantly different at P < 0.05 within each date of measurement.
Table 2. Diurnal inhibition of net photosynthetic rate after fruit removal at different stages of fruit development in field-grown ‘‘Golden Delicious’’ apple in 1990. Negative values indicate an increase in photosynthesis after fruit removal.
Date of fruit removal Inhibition of photosynthesis (%)
< 24 h AFR1 24--72 h AFR
Morning Afternoon Morning Afternoon
June 28 4 11 2 25
July 17 10 9 34 36
August 6 −5 4 17 25
September 10 3 3 6 12
to PPFDs < 400 µmol m−2 s−1. For example, Lakso and Seeley (1978) concluded that only 15--20% of the leaf area of apple trees is exposed to saturating PPFD, and Heinecke (1963) estimated that the percentage of leaf area exposed to 60--100% of full sunlight was 25% in 17-year-old ‘‘Golden Delicious’’ trees trained as open vases. Based on our observations on the inhibition of A, data on light distribution within the canopy (Heinecke 1963, Lakso and Seeley 1978), and the fact that leaves distant from fruits do not show appreciable inhibition of A, we hypothesize that the effect of fruit removal on leaf photosynthesis is attenuated at the canopy level, with minor or negligible consequences on total carbon assimilation by the tree. However, the depression of A due to defruiting will alter the overall carbon fixation and accumulation of small canopies (Monselise and Lenz 1980), shoots or spurs, and it should be considered for carbon budget models of shoots and small trees.
Acknowledgments
We thank Drs. E. Magnanini and F. Rossi for helping to determine the neutrality of filters used for A/Q response curves. Research supported by the National Research Council of Italy, special grant RAISA, Subproject No. 2, Paper No. 1935.
References
Barden, J.A. 1974. Net photosynthesis, dark respiration, specific leaf weight, and growth of young apple trees as influenced by light regime. J. Am. Soc. Hortic. Sci. 99:547--551.
DeJong, T.M. 1986. Fruit effects on photosynthesis in Prunus persica. Physiol. Plant. 66:149--153.
Ehleringer, J. and O. Bjorkman. 1977. Quantum yields for CO2 uptake in C3 and C4 plants. Dependence on temperature, CO2 and O2 concentrations. Plant Physiol. 59:86--90.
Ferree, D.C. and J.W. Palmer. 1982. Effect of spur defoliation and ringing during bloom on fruiting, fruit mineral level, and net photo-synthesis of ‘‘Golden Delicious’’ apple. J. Am. Soc. Hortic. Sci. 107:1182--1186.
Figure 3. Response of net photosynthetic rate (A) to varying photosyn-thetic photon flux density (PPFD) measured at 340--350 µl l−1 ambient CO2 between 1245 and 1400 h on (a) August 9, 1990, (b) Septem-ber 12, 1990 and (c) June 26, 1991. The horizontal bar indicates the PPFD range (0--200 µmol m−2 s−1) where apparent quantum yield was calculated (values reported in Table 3).
Table 3. Apparent quantum yield of leaves of fruiting and defruited field-grown ‘‘Golden Delicious’’ apple on M27 rootstock. Quantum yield was calculated as the slope of the regression line fitted to assimilation response curves to PPFD from 6--9 measurements in the 0--200 µmol m−2 s−1 range (R2 > 0.86). Net photosynthetic rate (A) was measured on 3--4 leaves per tree at 330--355 µl l−1 ambient CO2. Each data set corresponds to one tree.
Date Replicate Apparent quantum yield Adefruited /Acontrol 1
(mol CO2 mol−1 photon) (× 100)
Control Defruited2
August 9, 1990 1 0.014 0.013 98
2 0.014 0.013 58
3 0.016 0.021 57
September 10, 1990 1 0.017 --
--2 0.016 --
--September 12, 1990 1 0.016 0.014 83
June 26, 1991 1 0.024 0.021 92
2 0.020 0.019 68
3 0.018 0.017 80
Fujii, J.A. and R.A. Kennedy. 1985. Seasonal changes in the photosyn-thetic rate on apple trees: a comparison between fruiting and non-fruiting trees. Plant Physiol. 78:519--524.
Gucci, R. and J.A. Flore. 1992. Net photosynthesis and nonstructural carbohydrates in leaves of field-grown plum trees following fruit removal. In Regulation of Photosynthesis in Fruit Crops. Ed. A. Tombesi. Perugia, Italy, pp 91--96.
Gucci, R., C. Xiloyannis and J.A. Flore. 1991. Gas exchange parame-ters, water relations and carbohydrate partitioning in leaves of field-grown Prunus domestica L. following fruit removal. Physiol. Plant. 83:497--505.
Guinn, G. and J.R. Mauney. 1980. Analysis of CO2 assumption: feedback control. In Predicting Photosynthesis for Ecosystem Mod-els. Eds. J.D. Hesketh and J.W. Jones. CRC Press, Boca Raton, Florida, USA, pp 3--16.
Hansen, P. 1970. 14C-Studies on apple trees: VI. The influence of the fruit on photosynthesis of the leaves, and the relative photosynthetic yields of fruits and leaves. Physiol. Plant. 23:805--810.
Heinecke, D.R. 1963. The microclimate of fruit trees----II. Foliage and light distribution patterns in apple trees. Proc. Am. Soc. Hortic. Sci. 83:1--11.
Lakso, A.N. and E.J. Seeley. 1978. Environmentally induced re-sponses of apple tree photosynthesis. HortScience 13:646--650. Monselise, S.P. and F. Lenz. 1980. Effect of fruit load on
photosyn-thetic rate of budded apple trees. Gartenbauwissenschaft 45:220--224.
Preston, A.P. 1978. A bed system of planting apples on the dwarfing rootstock M27. Acta Hortic. 65:229--235.
Proctor, J.T.A., R.L. Watson and J.J. Landsberg. 1976. The carbon budget of a young apple tree. J. Am. Soc. Hortic. Sci. 101:579--582. Roper, T.R., J.D. Keller, W.H. Loescher and C.R. Rom. 1988. Photo-synthesis and carbohydrate partitioning in sweet cherry: fruiting effects. Physiol. Plant. 72:42--47.
Schaffer, B., L. Ramos and S.P. Lara. 1987. Effect of fruit removal on net gas exchange of avocado leaves. HortScience 22:925--927. Schechter, I., D.C. Elfving and J.T.A. Proctor. 1991. Apple tree canopy
photosynthesis as affected by rootstock. Can. J. Bot. 69:295--300. Steel, R.G. and J.H. Torrie. 1980. Principles and procedures of
