ABSTRACT
“Washington Navel” orange is the principal mid-season maturing cultivar grown in south-eastern Australia. The cultivar has strong export potential and earns high returns for citrus growers. However, trees bear a large number of small-sized fruit in an on-year followed by a small number of large-sized fruit the next year if crop load is not properly managed. Thinning can be used as a crop management strategy to adjust crop load. A thinning experiment was conducted, and crop loads were adjusted on a whole tree basis to 2, 4, 6 or 8 fruit/0.125 m3 canopy volume (CV), removing 46%, 30%, 20% and 8% of the crop respectively, and compared to non-thinned control trees. Fruit diameter was measured at regular intervals from February to July. Fruit yield and number were recorded for two consecutive seasons. Return bloom after thinning was also recorded. Thinning treatments 2, 4 and 6 fruit/0.125 m3 CV resulted in higher fruit growth rates of 24%, 9% and 14% compared to the control. Thinning to a crop load of 6 fruit/0.125 m3 CV produced the best results for fruit size and yield during the treatment year. The following year return bloom was increased in the 2, 4 and 6 fruit/0.125 m3 CV compared to the control. The 6 fruit/0.125 m3 CV treatment resulted in a higher cumulative yield of 169 kg/tree over two cropping seasons and 42% fruit in the size class > 75 mm diameter compared to the control. Economic analysis revealed that 6 fruit/0.125 m3 CV could generate an extra $4,240/ha of income for citrus growers. It was concluded that a fruit load of 6 fruit/0.125 m3 CV was required to produce high yields and large fruit size over two cropping seasons.
Introduction
Fruit size is one of the most important quality characteristics for selling Australian navel oranges on both domestic and export markets. Large fruit sell at higher prices, while small fruit are difficult to sell and there is limited demand for them in the juice industry. Fruit size is mainly determined by the genetic make-up of the cultivar but can be influenced by several other factors such as rootstock, cultural practices, tree size, crop load and climatic conditions experienced during fruit growth and development (Mechlia and Carroll, Citation1989; Reuther, Citation1973). Three growth stages have been identified during citrus fruit development: Stage I (cell division), Stage II (cell enlargement) and Stage III (maturation) (Bain, Citation1958). In navel orange, the transition from Stage I to Stage II has been reported to occur around mid-late December in the inland growing areas of south-eastern Australia (Bevington and Khurshid, Citation2002).
Fruit thinning can be used as a strategy to improve fruit size, yield and return flowering, and to reduce the alternate bearing cycle in some citrus cultivars (El-Sayed et al., Citation2017). Fruit thinning typically reduces yield in the year of thinning (Davis et al., Citation2004), but monetary returns from the larger fruit size can compensate for the thinning costs (McNeil et al., Citation1994). In the following year, the higher yields can compensate for yield losses and prevent the trees from going into an alternate bearing cycle (Guardiola and Garcia-Luis, Citation1997).
Hand thinning is a common practice in a range of fruit crops including citrus. Zaragoza et al. (Citation1992) successfully used hand thinning to reduce crop load by 66% and improved fruit diameter by 2.5 mm at harvest in “Satsuma” mandarin. Sutton and Harty (Citation1990) reported an increase in fruit size from reducing the crop load by 30–54% with no effect on yield in “Satsuma” mandarin. Rodrigues et al. (Citation1998) improved fruit quality in southern Brazil by hand thinning 66–83% of the crop in tangerines. The increase in the fruit size distribution in Valencia oranges in South Africa was achieved by reducing the crop load by 10–20% (Rabe, Citation1991). Hand thinning in “Imperial” mandarin reduced the crop load sufficiently to achieve an increase in fruit size, and at the same time, minimized any subsequent alternate bearing tendency the following year (Bevington et al., Citation1998).
One of the reasons for increased fruit size after thinning could be reduced competition among the remaining fruitlets, enabling better fruit development. While this is generally accompanied by yield reduction due to lower crop loads (Bustan et al., Citation1995; Guardiola and Garcıa-Luis, Citation2000), and increased fruit yield occurs in the following year.
Hand thinning is normally an expensive crop management practice, and the reduction of fruit yield would be a disadvantage if the increase in fruit size does not translate into increased market value. Hand thinning is not widely practiced by navel orange growers in Australia due to the cost of labor, and growers often do not have an accurate knowledge that how many fruit are required to thin to achieve the desired fruit size without compromising yield. However, growers have started to realize the benefits of hand thinning in the last few years. Selective hand thinning appears to be a very effective and accurate method for removing undersized and blemished fruit to adjust the crop load to a desired level in “Nadorcott” mandarins (Stander and Cronje, Citation2016). Currently, there appears to be no refereed published data describing the appropriate hand thinning strategies in relation to fruit size enhancement and yield increase in navel oranges.
The objective of this investigation was to examine the responses of “Washington Navel” trees to different levels of hand thinning to identify the optimum crop load for achieving an appropriate balance between enhanced fruit size and yields and to optimize net returns over two cropping seasons.
Materials and Methods
Site Description
A field experiment was conducted in a commercial orchard at Dareton (34° 1’ S, 141° 9’ E), NSW, Australia on “Washington Navel” [Citrus sinensis (L.) Osb.] for two growing seasons (2011–12 and 2012–13). The “Washington Navel”/Poncirus trifoliata trees were planted in 1978 at 3.4 m × 6.7 m spacing in east-west oriented rows. Tree density was 440 trees/ha with an average tree height of 2.5 m and the average canopy volume (CV) was 9 m3. The soil texture was sandy loam down to at least 140 cm. The active root zone was approximately 80–100 cm deep and soil EC ranged between 0–0.40 ds/m. The pH of the soil ranged between 8 and 9 (water). The typical dry semi-arid climate of south-eastern Australia necessitated scheduled irrigation and the trees received (12 ML/ha/year from low-level, under-tree sprinklers. Standard orchard management such as pest and disease control and nutrition application were conducted according to commercial standards.
The average daily maximum temperature at Dareton ranges from 30–32°C during December – February and 16–17°C during June – August, with an average annual rainfall of 220 mm. Annual accumulated heat units average 1,880 (calculated as [maximum temperature + minimum temperature]/2–13°C (maximum threshold of 35°C). The accumulated heat units in 2011/12 and 2012/13 for December – February were 836 and 1,071, and accumulated rainfall figures were 15 mm and 84 mm, respectively.
Fruit Thinning
Before thinning treatments started, fruit density counts were recorded on each experimental tree using a 0.5 m × 0.5 m × 0.5 m (0.125 m3) fruit counting frame as a guide. The frame was randomly placed around the tree canopy at six different positions to determine the existing crop loads (number of fruit/0.125 m3 CV). Fruit were hand thinned to the desired levels and the frame was used to ensure the desired level of thinning was achieved. Hand thinning treatments were applied on 30 January 2012 after the completion of December fruit drop (110 days after full bloom) in a “heavy-fruit set” year. During thinning, damaged and blemished fruit were removed followed by the removal of fruit <50 mm in diameter.
The five treatments used were thinning to 2, 4, 6 or 8 fruit/0.125 m3 CV, removing 46%, 30%, 20% and 8% fruit respectively. The non-thinned control trees had an average fruit density of 12 fruit/0.125 m3 CV (). Five treatments were randomly allocated to six single tree blocks in a complete block design.
Table 1. Total fruit removed, canopy volume and trunk circumference measured per tree at the time of thinning in “Washington Navel” orange in 2012.
Data Collection
Fruit diameter (mm) was measured in 2012 on a total of 300 fruit: 10 representative fruits from each experimental tree. Measurements were carried out fortnightly from 23 February to 1 July 2012 using digital calipers (Mitutoyo Corp., Japan) with a precision of ±0.1 mm. Average fruit diameter was 58 mm at the first measurement. Fruits were harvested on 6 July 2012 and 10 July 2013 at commercial maturity. Fruit yield and the total number of fruits per tree were recorded for each experimental tree in both seasons. The fruit was graded into different size classes with a commercial grader [Colour Vision Systems Pty Ltd., Australia (MAF Roda Group)] to determine the fruit size distribution at harvest. Flowering data were recorded on 10 October 2012 the year following thinning. The number of flowers, leafless and leafy inflorescences, and vegetative shoots were counted/0.125 m3 CV using a counting frame at six different positions around the tree for a total of 30 trees.
Statistical Analysis
Statistical analyses were performed with GenStat (VSN International, Citation2022) statistical software package using a one-way analysis of variance (ANOVA) in randomized blocks with a single treatment factor (thinning). Second order polynomials were fitted to the fruit growth data and the growth curve parameters were analyzed with ANOVA to assess the true difference across treatments. Where the F-test demonstrated significant treatment effects, means were separated using least significant difference (LSD) calculated at p = .05. Regression analyses were also used to assess the relationship between crop loads, final fruit size and return bloom.
Results
Season 2011–12
Tree vegetative characteristics such as canopy volume and trunk circumference were uniform for the different thinning treatments at the start of the experiment in 2011–12 ().
Fruit Growth
Seasonal fruit growth patterns showed increased size following thinning. Second order polynomial growth curves were fitted to the data sets. These curves explained 99% of the total variability in the data throughout the fruit measurement period (). Analysis of the curve parameters indicated significant differences between thinning treatments in linear components (). The initial fruit diameter of the 2, 4 or 6 fruit/CV treatments was significantly higher than the 8 fruit/CV or control. Fruit growth rates were higher on trees with the heaviest thinning treatments compared to non-thinned or those thinned to 8 fruit/CV. Fruit from the 2 fruit/CV thinning treatment showed a higher growth rate than all other treatments and, considering the very low crop load on these trees, would reflect growth potential under non-limiting resource conditions. Fruit diameter slope was not affected by any treatment (). Total fruit growth increments from February to July were 24%, 9% and 14% higher for 2, 4 and 6 fruit/CV treatments respectively, compared to non-thinned trees (). The lightest thinning treatment of 8 fruit/CV was ineffective in enhancing fruit growth.
Figure 1. Fruit growth patterns in “Washington Navel” trees for different thinning treatments 2, 4, 6 or 8 fruit/0.125 m3 canopy volume (CV) or non-thinned (control) during the 2012 cropping season.
Table 2. Parameters of the quadratic curve fitted to the fruit diameter data recorded 23 February–1 July 2012 for thinned and non-thinned “Washington Navel” orange trees.
Fruit Size Distribution and Yield
In 2012, the non-thinned (control) trees produced only 93 fruit/tree in the preferred large size range (>75 mm); the rest of the fruit was in the undesirable small size range (). The 2, 4 and 6 fruit/CV thinning treatments significantly increased the number and proportion of large fruit produced compared to the control. These treatments also resulted in a substantial reduction in the number and proportion of small fruit produced, which in part resulted from the selective removal of smaller fruit during thinning. There was a linear relationship (R2 = 0.62) between crop load and the proportion of large fruit produced at harvest (). Linear regression analysis identified fruit loads associated with final fruit size. Significant reductions in yields were only evident in the 2 and 4 fruit/CV thinning treatments where yields were reduced by 25% and 34%, respectively compared to non-thinned trees (). On the other hand, yields were increased in 6 fruit/CV treatments by 25% and 42% compared to 2 and 4 fruit/CV, respectively. The total number of fruits in 2012 followed a similar trend to fruit yield data ().
Figure 2. The relationship of (a) fruit number/m3 canopy volume (CV) vs. percent fruit >75 mm/tree produced after thinning; and (b) fruit number/tree vs. total flower number/0.125 m3 CV the year after thinning.
Table 3. Fruit number (>75 mm)/tree, yield/tree and fruit number/tree for thinned and non-thinned “Washington Navel” trees for 2012 and 2013 seasons.
Net Returns
Net returns were calculated from average market prices for each size class at the time of harvest in 2012. Net returns per hectare (after allowing for harvesting and packing costs) were highest for the 6 fruit/CV thinning treatment ($18,380) compared to $14,140 for non-thinned trees, due to the higher per carton returns for large fruit ().
Table 4. Effect of hand thinning on per carton returns and net returns per hectare of “Washington navel” orange in thinned and non-thinned trees in 2012.
Net returns for the 2 and 4 fruit/CV thinning treatments were also higher than returns from non-thinned trees; however, despite higher average carton prices, returns were lower for 2 and 4 fruit/CV, due to lower yields in 2012 compared to 6 fruit/CV thinning treatment. Taking thinning costs into account, the 2, 4 and 6 fruit/CV thinning treatments all resulted in higher net income compared to non-thinned trees. The lightest thinning treatment 8 fruit/CV resulted in a net loss of income ().
Season 2012–13
Return Flowering
Flower counts were undertaken on 15 October 2012 to assess previous thinning treatment effects on return bloom and its effects on the potential yield and fruit size in the second year following thinning. Trees previously thinned to 2, 4 or 6 fruit/CV produced significantly more flowers and leafy inflorescences than non-thinned trees or trees thinned to 8 fruit/CV (). Apart from the control, all thinning treatments reduced the number of vegetative shoots per CV. Linear regression analysis was used to estimate fruit loads associated with target return flowering (). The lower fruit loads were associated with the higher flower numbers and vice versa; however, the intermediate fruit loads resulted in acceptable flower numbers for the next year’s crop.
Table 5. Effects of hand thinning on return flowering/0.125 m3 canopy volume for “Washington Navel” orange in 2012–13 season.
Fruit Size Distribution and Yield
In 2013, the number of large fruit (>75 mm) was significantly higher in the 2 or 6 fruit/CV treatment compared to all other treatments including control trees in the year following thinning (). The number of fruit >75 mm in the control treatment increased but remained lower than the 2 or 6 fruit/CV treatments. There was a significant increase in fruit yield with the previous thinning treatment of 2 fruit/CV compared to the control (). The results suggest that there was no decrease in fruit yield with the 6 fruit/CV treatment. The total number of fruit/tree in 2013 followed a similar trend to fruit yield data ().
Discussion
Fruit thinning is a common practice for increasing fruit size in commercial fruit production. The effect of fruit thinning on fruit size is directly related to the large-sized fruit left behind on the trees and less competition between the remaining fruit for available water, nutrients and carbohydrates. Previous crop load and optimum climatic conditions also play a vital role in fruit size.
The greater number of fruits produced by a tree in an “on-flowering” year will be smaller sized at harvest with lower economic benefits to the grower. Conversely, fruit removal causes increases in the average weight, or size, of fruit at harvest. However, at some point removing additional fruit will not increase fruit size as the genetic potential of the crop has been achieved (Bevington and Khurshid, Citation2002). The problems that citrus growers are facing include how to thin their fruit, how much fruit to thin and when to thin without significant losses in yields. Higher or acceptable yields should also coincide with the desired fruit size, which is suitable for export to lucrative markets at higher net returns.
Early indications of potential fruit size problems would enable growers to remove smaller fruit early in the season to enhance the growth rate of the fruit left on the tree. Timing of thinning is therefore important and the timely removal of small fruit to decrease crop load would allow greater partitioning of reserves into the remaining fruits with possible gains in fruit size (Goldschmidt et al., Citation1992). Fruit thinning in this study was commenced after the complete natural fruit drop period. Generally, fruit growth rate does not fluctuate after the complete fruit drop period and from there onward, growth rate stability improves with time. Therefore, it was assumed that an accurate fruit size adjustment would be possible after the completion of Stage I fruit development. Fruit seems to be less responsive to thinning once past a period of linear growth, which varies with the crop load (Palmer et al., Citation1997), and unfavorable climatic conditions (Mechlia and Carroll, Citation1989).
Fruit growth was monitored during the entire growth period across different thinning treatments until harvest. The fruit growth curves were similar to those described for citrus by Franco and Gravina (Citation2000) and Zhang et al. (Citation1992); however, fruit growth was affected by various crop loads. The effect of crop load on fruit growth was previously reported by Palmer et al. (Citation1997) and Buwalda et al. (Citation1989). The growth curve parameters indicated that initial fruit size is of vital importance and its effect carries through to harvest. This suggests that the potential for fruit size is determined at a very early growth stage in the season. Large fruit at the beginning of the cell enlargement period become larger fruit at harvest, confirming reports on “Washington Navel” (Khurshid and Braysher, Citation2009), “Clementine” mandarin (Koch et al., Citation1996) and apples (Stanley et al., Citation2001).
Fruit yields were significantly lower in heavily thinned trees with the removal of 30–46% of the crop. Yield reduction at such a low crop load was expected. Similar results of low yields after thinning have been reported (Guardiola and Garcia-Luis, Citation1997). However, the removal of 20% of the crop had a reasonable increase in fruit yield over the two heavily thinned treatments. The following year there was no reduction in yields with the previously 6 fruit/CV treatment and trees were able to produce a high proportion of large-sized fruit. The 2 fruit/CV treatment had increased yields and enhanced fruit size due to the low crop load in the previous year. Generally, fruit size was large in the year following thinning. This was also partly due to the warm temperatures experienced during the active fruit growth period from December – February supplemented by the 84 mm of rainfall in 2013 compared to 35 mm in 2012, which would have contributed to the large fruit size at the cell enlargement period. This agrees with our previous study on fruit growth of “Washington Navel” oranges (Khurshid and Braysher, Citation2009).
The removal of 46% and 30% of the crop in the previous year caused an alternate bearing pattern in yield over two years, while 20% removal produced uniform yield over two years. Regression analysis suggested that the previous crop load had a stronger effect (R2 = 0.62) on final fruit size at harvest, while the rest of the variation could be due to management practices and overriding climatic conditions.
The effect of crop load on return bloom was significant. Regression analysis explained 59% of variation in the data. The large number of fruit >75 mm (diameter) in 2013 for the 2, 4 or 6 fruit/CV treatments was due to the increased number of flowers and leafy inflorescences. Fruit set on leafy inflorescences usually attain a large size (Davenport, Citation1990; Khurshid and Krajewski, Citation2010; Krajewski and Rabe, Citation1995). The large number of leafy inflorescences in the 2, 4 or 6 fruit/CV treatments could partly be responsible for the enhanced fruit size at harvest.
The economic analysis suggested strong differences between the net returns from large and small-sized fruit. However, in relation to fruit size, optimum crop load will depend on the prevailing price differential between large and small fruit size counts. Although the 2 or 4 fruit/CV thinning treatments resulted in the highest proportion of large fruit, overall yield was not high enough to produce the highest net returns after accounting for thinning costs. For trees at this site, an intermediate crop load (6 fruit/CV) produced the highest net returns in a heavy crop year and the following year.
These results obtained over two years indicated that navel orange trees carrying a heavy crop load can be economically hand thinned. Although fruit size is largely determined during Stage I, it was further indicated that some enhancement of fruit growth rates in heavy crop years can be obtained by reducing crop loads during early Stage II of fruit development. To achieve a meaningful increase in growth rate, a reduction of 20–45% in crop load is likely to be required. Medium-sized trees with a 25 cm2 circumference used in this study had the potential to grow up to 400 fruit at the same growth rate in the year of thinning and the following year with optimum management and climatic conditions. These results clearly suggested that moderate fruit thinning in an “on-year” has greater benefits in the year of thinning and the following year without the need for any further thinning. The yield and fruit size in two years offset the cost of thinning in the previous year.
Conclusion
Hand thinning is not often applied as a commercial cultural practice in sweet oranges due to the practice’s reliance on costly manual labor. However, hand thinning could provide unique benefits such as treatment selectivity and easier control over thinning intensity. A fruit thinning experiment was conducted in the “on-flowering” year after natural fruit drop was completed. Leaving large fruit on trees improved fruit size at harvest. Fruit thinning to 6 fruit/CV increased income, enhanced the number of fruits in the export size class (>75 mm) and did not cause any yield losses. Fruit thinning had a positive effect on the next year’s crop and eliminated the alternate bearing cycle and net returns per hectare were highest. Fruit load adjustment of 6 fruit/CV is recommended in an expected “on-flowering” year.
Disclosure Statement
No potential conflict of interest was reported by the author(s).
Additional information
Funding
References
- Bain, J.M. 1958. Morphological, anatomical and physiological changes in the developing fruit of the Valencia orange, Citrus sinensis (L.) Osbeck. Aust. J. Bot. 6(1):1–23. doi: 10.1071/BT9580001.
- Bevington, K.B., P. Florissen, S. Gee, and S. Falivene, 1998. Improving ‘imperial’ mandarin fruit quality and marketability. Final report project CT95031, Horticulture Australia Limited, Sydney.
- Bevington, K.B., and T. Khurshid, 2002. Optimisation of citrus production and fruit size: An interactive management model. Final report project CT98023, Horticulture Australia Limited, Sydney.
- Bustan, A., Y. Erner, and E.E. Goldschmidt. 1995. Interactions between developing citrus fruits and their supportive vascular system. Ann. Bot. 76(6):657–666. doi: 10.1006/anbo.1995.1144.
- Buwalda, J.G., D.J. Klinac, and J.S. Meekings. 1989. Effects of time and degree of fruit thinning on fruit size and crop yield at harvest for four Nashi (Pyrus serotina Rehd.) cultivars. Sci. Hortic. 39(2):131–141. doi: 10.1016/0304-4238(89)90085-X.
- Davenport, T.L. 1990. Citrus flowering, In: pp. 349–408. In: J. Janick (ed.). Hortic. Reviews. Vol. 12. Timber Press, Portland, Ore. USA.
- Davis, K., E. Stover, and F. Wirth. 2004. Economics of fruit thinning: A review focusing on apple and citrus. HortTechnology 14(2):282–289. doi: 10.21273/HORTTECH.14.2.0282.
- El-Sayed, F.S., H.S. Ahmed, and A.M. Elazazy. 2017. Effect of hand and chemical thinning on regulating alternative bearing in ‘Murcott’ trees. J. Hortic. Sci. Ornam. Plants 9:98–105.
- Franco, J.E., and A. Gravina. 2000. Early prediction of fruit size at harvest in ‘Valencia’ orange Citrus sinensis (L) osb. using experimental data. Proc. Intl. Soc. Citric. 1:456–458.
- Goldschmidt, E.E., A. Harpaz, S. Gal, D. Rabber, and E. Gelb. 1992. Simulation of fruitlet thinning effects in citrus by a dynamic growth model. Proc. Int. Soc. Citric. 1:515–519.
- Guardiola, J.L., and A. Garcia-Luis. 1997. Thinning effects on citrus yield and fruit size. Acta Hortic. 463(463):463–474. doi: 10.17660/ActaHortic.1998.463.58.
- Guardiola, J.L., and A. Garcıa-Luis. 2000. Increasing fruit size in citrus. Thinning and stimulation of fruit growth. Plant Growth Regulat. 31(1):121–132. doi: 10.1023/A:1006339721880.
- Khurshid, T., and B. Braysher. 2009. Early fruit size prediction model using cubic smoothing splines for ‘Washington Navel’ (Citrus sinensis L. Osbeck) oranges in Australia. Int. J. Fruit Sci. 9(4):394–408. doi: 10.1080/15538360903378724.
- Khurshid, T., and A. Krajewski. 2010. Bearing branch units developed on branches hedged during flowering produce large ‘Washington Navel’ (Citrus Sinensis L. Osbeck) oranges. Int. J. Fruit Sci. 10(3):215–227. doi: 10.1080/15538362.2010.510417.
- Koch, N., K. Theron, and E. Rabe. 1996. Fruit size prediction in ‘Clementine’ and ‘satsuma’ mandarin and improvement of ‘Clementine’ mandarin (citrus reticulata Blanco) fruit size by 2,4-DP. Proc. Int. Soc. Citric 2:737–744.
- Krajewski, A., and E. Rabe. 1995. Citrus flowering: a critical evaluation. J. Hort. Sci 70(3):357–374. doi: 10.1080/14620316.1995.11515306.
- McNeil, R.J., K.C. Johnston, K.W. Johnson, and R.W. Smith. 1994. A demonstration of the economics and effectiveness of 1-naphthaleneacetic acid (NAA) for fruit thinning of ‘Valencia’ orange. Proc. 6th Ann. Conf. Western Plant Growth Regula. Soc. 21–33.
- Mechlia, N.B., and J.J. Carroll. 1989. Agroclimatic modelling for the simulation of phenology, yield and quality of crop production. II. Citrus model implementation and verification. Int. J. Biometerol. 33(1):52–65. doi: 10.1007/BF01045897.
- Palmer, J.W., R. Giuliani, and H.M. Adams. 1997. Effect of crop load on fruiting and leaf photosynthesis of ‘Braeburn’/M26 apple trees. Tree Physiol. 17(11):741–746. doi: 10.1093/treephys/17.11.741.
- Rabe, E. 1991. Fruit size improvement in ‘Valencia’ and ‘tomango’ orange cultivars [Citrus sinensis (L.) osbeck] through hand-thinning of fruitlets in November–December. J. S. Afr. Soc. Hortic. Sci. 1:59–63.
- Reuther, W. 1973. Climate and citrus behaviour, p. 280–337. In: W. Reuther (ed.). The citrus industry. Vol. 2. University of California, Division of Agricultural Sciences, Berkeley, United States.
- Rodrigues, L.R., S.F. Schwarz, V.P. Reckziegel, and O.C. Koller. 1998. Manual thinning of tangerine ‘Montenegrina’. Pesqui. Agropecu. Bras. 33:1315–1320.
- Stander, O.P., and P.J. Cronje. 2016. Reviewing the commercial potential of hand thinning in citrus with a cost-benefit analysis of summer hand thinning of ‘Nadorcott’. Mandarin. HortTechnology 26(2):206–212. doi: 10.21273/HORTTECH.26.2.206.
- Stanley, C.J., J.R. Stokes, and D.S. Tustin. 2001. Early prediction of apple fruit size using environmental indicators. Acta Hortic. 557(557):441–446. doi: 10.17660/ActaHortic.2001.557.59.
- Sutton, P.G., and A.R. Harty. 1990. Hand thinning satsuma mandarins pays off. Orchardist N.Z. 63(10):18–20.
- VSN International. 2022. Genstat for windows. 22nd ed. VSN International, Hemel Hempstead, UK.
- Zaragoza, S., I. Trenor, E. Alonso, E. Primo-Millo, and M. Augusti. 1992. Treatments to increase the final fruit size on Satsuma Clausellina. Proc. Int. Soc. Citric. 2:725–728.
- Zhang, L., G. Shen, J. Zhang, and H. Chen. 1992. Study on the law of orange fruit growth and development and the influence of meteorological factors. Proc. Int. Soc. Citric. 1:432–434.