25 1. Introduction Cherry fruit size and quality is an important factor in production and sales of sweet cherry fruit (Proebsting and Mills, 1981). Sweet cherry trees are typically upright, vig- orous and non-precocious (Lang et al., 2004) so orchard management practices focus on achieving high yields of premium quality fruit through balancing reproductive and vegetative growth. Manipulation of the number of fruits (crop load) on the tree, and leaf area, can be used to en- courage larger and sweeter fruit through balanced carbo- hydrate supply and demand (Lang et al., 2004; Spayd et al., 1986; Whiting and Lang, 2004). However, in many of these studies, yield losses due to cracking have not been presented (Proebsting and Mills, 1981) even when the economic losses due to cracking can be significant (Han- son and Proebsting, 1996). Given that cracking can be in- duced by internal vascular flow (Measham et al., 2009), it is posited that higher crop loads will reduce the incidence of cracking through increased competition between fruit for assimilate supply. It has been previously hypothesised that higher crop loads increase competition between fruit for carbohy- drates and that lower crop loads result in higher assimi- late supply for individual fruit (Spayd et al., 1986), and that there can be a resultant increase in size (Spayd et al., 1986) and concentration of sugars (Proebsting and Mills, 1981). It has also been noted however, that lower crop loads are associated with increased vegetative growth (Kappel, 1991) and that current season’s vegetative growth is a strong sink for carbohydrates. Diurnal trans- location of sugars from leaves to fruit can be variable (Richardson, 1998), and therefore it is difficult to assess relationships between sugars and cracking as a result of internally supplied excess water. Cherry fruits are strong sinks (Ayala and Lang, 2008) and it has been noted that removal of spur leaves had lit- tle effect on fruit quality because alternative supplies of carbohydrates were sourced (Whiting and Lang, 2004). Fruit and leaf ratio can be manipulated for optimum qual- ity. Two flower buds per spur has been suggested as the ideal (Whiting and Lang, 2004). An interaction between fruit and leaves was also implicated in the development of cracking, in a study by Measham et al. (2010), which showed that leaf removal decreased the development of side cracks in cherry fruit during the few weeks prior to harvest. Furthermore, diurnal water potential gradients and evaporative demands on the leaf influenced vascular flow to the fruit demonstrating a local fruit and leaf interaction. Thus, given that fruit size (Simon, 2006) and sugar lev- els (Christensen, 1996) have been associated with the de- velopment of cracking, and Simon (2006) cites two studies (Bullock, 1952; Way, 1967) that found trees with high loads that showed little cracking within variety, the potential for crop load manipulation to influence fruit cracking warrants investigation. The aim of this present study is to further ex- amine this relationship between crop load and cracking. Crop load manipulation and fruit cracking in sweet cherry (Prunus avium L.) P.F. Measham*, S.A. Bound**, A.J. Gracie*, S.J. Wilson* * Tasmanian Institute of Agriculture, University of Tasmania, Private Bag 98, Hobart, Tasmania, 7001 Australia. ** Tasmanian Institute of Agriculture, 13 St Johns Avenue, New Town, Tasmania, 7008, Australia. Key words: fruit crop load, incidence of cracking, sweet cherry. Abstract: Yield loss from rain-induced fruit cracking is a perpetual risk associated with the production of sweet cherries, and is difficult to manage due to the unpredictability of fruit responses to late season rainfall. The aim of this five-year study was to investigate the relationship between fruit crop load and incidence of cracking. The results showed a negative correlation between crop load and incidence of fruit cracking, and it was found in both natural and manipulated crop load trials for all varieties studied and in all seasons assessed. The effect of crop load on final cracking levels are deter- mined post cell division. Results from this study showed that fruit width was positively correlated with cuticular cracking but, contrary to what has been purported in literature, no relationship between concentration of soluble sugars or firm- ness with the incidence of cracking was found. This study has confirmed that crop load should be a major consideration in orchard practices in developing strategies to manage fruit cracking. Adv. Hort. Sci., 2012 26(1): 25-31 Received for publication 23 October 2011 Accepted for publication 3 February 2012 26 2. Materials and Methods Plant material Mature trees, grown on F12/1 rootstock, were used in all field trials. Trials were undertaken from late October to late January during seasons 2005/06, 2006/07, 2007/08 and 2010/11 in two commercial orchards in Huonville and Old Beach, Tasmania (Australia). All orchards were subjected to standard industry management practices. To investigate the effect of crop load on fruit cracking and type, five manipulated crop load trials were undertaken in years with late summer rainfall; Trials 1 and 2 in sea- sons 2005/06 and 2006/07 respectively, Trials 3, 4 and 5 in season 2010/11. A study of fruit properties from Trial 5 was undertaken. In addition, a survey of natural crop load and fruit properties over three years (2005/06, 2006/07, 2007/08) was performed. The relationship between levels of cracking in situ and the cracking potential using the cracking index (Christensen, 1972) was also evaluated. Manipulated crop load trials To assess the impact of crop load on crack develop- ment, manipulated crop load trials were undertaken on one variety ‘Simone’ at one site (Huonville) in two seasons: 2005/06 (Trial 1) and 2006/07 (Trial 2). Three further manipulated crop load trials were undertaken on differ- ent varieties and sites in one season, 2010/11; on variety ‘Sweetheart’ at Huonville and Old Beach (Trial 3 and 4 respectively), and on variety ‘Regina’ at Huonville (Tri- al 5). These varieties were chosen due to the variety of crack types they had previously displayed in earlier stud- ies (Measham et al., 2009) in Tasmania; ‘Simone’ showed a tendency for cuticular cracks, ‘Regina’ for side cracks’ and ‘Sweetheart’ for both. In Trials 1 and 2, treatments included a low, medium and high crop load, which aimed for 2, 5 or 8 fruit per cm2 trunk cross sectional area (TCSA) respectively in a randomised complete block design with three replicates (whole tree plots). Treatments were applied at pit-harden- ing during stage II of fruit growth and development which occurred post bloom at 4 weeks after full bloom (4WAFB). Where the high crop load specified could not be reached, natural crop load was determined and used. In Trials 3, 4 and 5, treatments included a low, medium and high crop load, applied at three different growth stages in a factorial design with five replicates (whole tree plots). Crop load was achieved by thinning each bud to 1, 2 or 4 floral buds per spur and applied pre bloom (PrB) at dor- mant bud stage, full bloom (FB) and post bloom (PoB) at four weeks after full bloom. In addition, for Trial 3, a sub sample of 30 non-cracked blemish-free fruits were ran- domly selected from each replicate tree for individual fruit assessments for size, total soluble solids, and firmness. Mean fruit property values were used to assess relation- ships with crop load, and with the incidence of cracking in situ. In all trials cracking incidence was determined at har- vest. Fruits from each treatment were additionally as- sessed for cracking index (using 50 non cracked fruits per variety). In all manipulated crop load trials, the actual crop load achieved for all trial trees was recorded. Natural crop load and fruit properties survey Natural crop load was recorded at harvest over three seasons on three randomly selected whole trees of avail- able varieties which included ‘Kordia’, ‘Lapin’, ‘Regina’, ‘Simone’, ‘Sweetheart’, ‘Sylvia’ and ‘Van’. All fruits were harvested and cracking levels recorded. Cracking incidence recorded at harvest was assessed in relation to natural crop load. In addition, for each season non-cracked blemish-free fruit from each variety was grouped, and a sub sample of 30 non-cracked fruits were selected for individual fruit as- sessments for size, weight, total soluble solids, stem length and skin thickness. Mean fruit property values were used to assess the relationship with the incidence of cracking in situ for each variety. Also in 2007/08, fruits from available varieties were assessed for cracking with the cracking index immersion method (Christensen, 1972). The relationship between the cracking index and the level of cracking recorded in the field was investigated. Given that this immersion method depends on water uptake across the skin, the im- pact of stem removal on uptake and the subsequent index value was also investigated using additional sub samples from variety ‘Simone’ with stems either removed or left intact. Measurements Cracking incidence was determined as per Measham et al. (2009), but with apical-end cracks and stem-end cracks combined to give a level of cuticular cracks. Measham et al. (2010) concluded that these crack types were likely to be induced through the same mode of water uptake. Crack- ing index was determined using the immersion method developed by Verner and Blodget (1931) as cited in and refined by Christensen (1972). All fruit were harvested between 7 a.m. and 12 noon and cracking assessments, morphological measurements and laboratory-based measurements were undertaken on the same day as harvest. Climate data for the months pre- ceding and during harvest was obtained from the Austra- lian Bureau of Meteorology Stations at Huonville (situated less than 5 km from the trial site) and at Old Beach using a PM-K208 PM-11 Phytomonitor Weather Station. Determination of crop load Prior to treatment application in manipulated crop load trials, tree girth was measured in centimetres at a point 5 cm above the graft union. TCSA was calculated for each tree for the area (A) of a circle using the formula (A = C2/4π), where C = circumference (cm) as described in Measham et al. (2009). Crop load was determined as total fruit number per TCSA. To determine natural crop load, all fruits from each tree were counted and crop load expressed as number of fruit per TCSA. 27 Fruit property tests Fruit size, weight and total soluble solid (TSS) concen- tration (brixo) were measured as described in Measham et al. (2009). Fruit firmness was measured using a Bioworks Inc. Firmtech 2 with values recorded using ControlSoft software. Stem length (mm) was measured using Vernier callipers and skin thickness was recorded microscopically using a Nikon SMZ800 dissecting microscope. Statistical analyses To assess the relationship between crop load and crack- ing incidence, and crop load with mean fruit properties, data were subject to linear regression tests and ANOVA. Interactions between crop load and timing of thinning to desired load were determined prior to assessing main ef- fects. Analysis of proportion data was performed on trans- formed data in order to meet the assumptions of analysis. To assess the effect of natural crop load on fruit proper- ties after accounting for variety, mean fruit property data were subject to ANOVA using variety as a fixed factor, and then to ANCOVA (crop load as the covariate) using PROC GLM (SPSS version 17). Unless specified, all results iden- tified as ‘significant’ are at probability level of 0.05. 3. Results Manipulated crop load trials All manipulated crop load trials received rainfall in the three weeks prior to harvest. In 2005/06 and 2006/07 there was a similar amount of rainfall (37 mm and 41 mm re- spectively). Trials 3 and 4 in 2010/11 experienced 49 mm and 50 mm rainfall respectively and Trial 5 received 42 mm rainfall (differing due to harvest dates). A negative linear relationship between actual crop load and total cracking incidence was recorded for variety ‘Simone’ over both seasons (2005/06 and 2006/07) (Fig. 1). The effect was greater in season 2005/06 than 2006/07, as indicated by the significantly greater magnitude of the slope for each crack type (Slope (B) = -4.69 and -0.77 for total and side cracks and -0.69 and -0.30 for cuticular cracks). In 2010/11 a significant interaction (P = 0.045) be- tween level and timing of crop load on total cracking in variety ‘Sweetheart’ at Huonville (Trial 3) (Fig. 2), but not at Old Beach (Trial 4). At Huonville (Trial 3), within thin- ning times, no significant effect of crop load at the dor- mant (PrB) or full bloom (FB) thinning times was seen, but there was a significant effect on total (P = 0.025) and side (P = 0.029) cracks (but not on cuticular cracks) when Fig. 1 - Percentage of total cracked fruit (A), cuticular-cracked fruit (B) and side-cracked fruit (C) with actual crop load (TCSA) for variety ‘Simone’. Each point is for an individual tree. A significant relationship was found between actual crop load and total cracking incidence in 2005/06 (R2 = 0.903, P<0.001) and 2006/07 (R2 = 0.511, P = 0.03), cuticular cracking 2005/06 (R2 = 0.907, P<0.001) and 2006/07 (R2 = 0.540, P = 0.02) and side cracking in 2006/07 (R2 = 0.575, P = 0.02). A significant difference between seasons was found for total cracking and cuticular cracking. Slope (B) of the linear regressions calculated for crop load and cracking incidence were significantly different between years. Fig. 2 - Incidence of total cracking at Huonville for variety ‘Sweet- heart’ under low, medium and high crop loads applied at dor- mant bud stage (PB), full bloom (FB) and four weeks after full bloom (PoB). A significant interaction between crop load level and application timing was evident; low crop load resulted in a significantly higher level of cracking within the post bloom application time only. 28 thinning was applied post bloom (PoB) (Table 1). High and medium crop load levels set by post bloom thinning resulted in significantly fewer cracked fruit than low crop loads. At Old Beach (Trial 4), there was no interaction be- tween crop load level and timing of thinning; there was, however, a significant main effect of crop load on both total (P = 0.01) and side (P = 0.01) cracks, but not on cu- ticular cracks. A significant main effect of crop load on total (P < 0.001) and side (P < 0.001) cracks was also seen in variety ‘Regina’ at Huonville (Trial 5) (Fig. 3). There was an interaction of crop load level and timing of thinning on fruit size in ‘Regina’ (Fig. 4) where low crop loads resulted in smaller fruit when thinned pre bloom or post bloom. Thinning at full bloom gave smaller fruit with medium loads. No interaction was found for fruit soluble solids or firmness; furthermore, no main effect of crop load or timing was found for fruit firmness. There was a significant main effect of crop load (P = 0.008) and timing of thinning (P = 0.013) on soluble solids with fruit from medium crop loads, and post bloom thinning displaying the highest soluble solids. No fruit properties were corre- lated with cracking levels except for fruit size, which was positively correlated (P = 0.01) with cuticular cracks only. Cracking indices determined for all treatments in Tri- als 3, 4 and 5 are given in Table 2. There was a significant relationship (P = 0.017) between index and total cracking Fig. 3 - Percentage of total cracked fruit (A) (R2 = 0.69) and side-cracked fruit (B) (R2 = 0.59) with actual crop load (TCSA) for variety ‘Regina’ and percentage of total cracked fruit (C) (R2 = 0.33) and side-cracked fruit (D) (R2 = 0.32) with actual crop load (TCSA) for variety ‘Sweetheart’. Each point is for an individual tree. Significant relationships were found between crop load and total cracking incidence and between crop load and side cracking. Table 1 - Incidence of total and side cracking in ‘Sweetheart’ at Hu- onville under high, medium and low crop loads applied post bloom Crack type Crop load Cracking incidence Total High 15.22 a Medium 16.62 a Low 27.40 b Side High 11.30 a Medium 12.44 a Low 22.89 b For each crack type, values followed by different letters indicate a sig- nificant difference (P<0.05). 29 recorded in situ for Trail 3 only; no relationship was evi- dent between cracking index and cracking in situ for Trial 4 or 5. Natural crop load and fruit property trials Lower natural crop loads had higher levels of crack- ing incidence (Fig. 5). Cracking incidence remained low (less than 5%) for crop loads higher than ten fruit per cm2 TCSA in all years and for all crack types (Fig. 5). When only using data points of less than 10 fruit per cm2 TCSA, relationships between cracking and crop load were found to be significant for all crack types in 2005/06 (total, R2 = 0.893, P < 0.001; cuticular, R2 = 0.853, P < 0.001; side, R2 = 0.540, P = 0.006) and for total and cuticular cracks in 2006/07 (total, R2 = 0.576, P = 0.005, cuticular, R2 = 0.463, P = 0.03). Across all varieties and seasons no significant relation- ship was found between any of the fruit property values with total cracking incidence or individual crack type in- cidence. Little difference can be observed between either weight or total soluble solids and changes in crop load, except perhaps a slight trend in variety ‘Sylvia’ where a decrease in weight, but not in total soluble solids, occurs with a dramatic increase (three fold to 24 fruit per TCSA) Table 2 - Cracking index (n = 50) determined for fruit from the three manipulated crop load trials (Trials 3, 4 and 5) Time of Application Crop Load Cracking in situ (%) Cracking index Trial 3 Trial 4 Trial 5 Trial 3 Trial 4 Trial 5 Dormant High 18 38 7 5 4 24 Medium 18 32 11 10 6 16 Low 14 59 28 11 15 21 Full Bloom High 14 31 6 8 8 3 Medium 18 39 11 9 7 7 Low 16 69 26 8 14 7 4WAFB High 15 28 7 6 9 4 Medium 17 32 11 9 4 7 Low 27 50 26 18 24 12 Mean incidence of total cracking is also given. A significant correlation was found between cracking index and total cracking in Trial 3 only. Fig. 4 - Fruit size (mm) determined for fruit from variety ‘Regina’ un- der low, medium and high crop loads applied at dormant bud stage (PrB), full bloom (FB) and four weeks after full bloom (PoB). A significant interaction between crop load and thinning time was found. Fig. 5 - The percentage of total cracked fruit (A), cuticular-cracked fruit (B) and side-cracked fruit (C) with natural crop load (TCSA). Each point is for an individual tree. A significant relationship was found between crop load and total cracking incidence in 2005/06 and 2006/07, and between crop load and cuticular cracking in 2005/06, and between crop load and side cracking in 2005/06 and 2006/07. 30 in crop load in the third season, 2007/08 (Fig. 6). A signifi- cant variety effect was found in fruit weight (P = 0.02), to- tal soluble solids (P = 0.001) and stem length (P < 0.001). After accounting for variety, and analysing data using crop load as a covariate, a significant effect of crop load was found for fruit weight only (P = 0.03). Fig. 6 - Mean fruit weight and total soluble solids for seven variet- ies over three seasons (error bars represent SEM, n = 30) with mean crop load levels (n = 3) and cracking incidence for the same varieties and seasons. For cracking incidence, expressed as percentage of total yield, columns represent total incidence, where light areas represent incidence of cuticular-cracked fruit, and dark areas represent incidence of side-cracked fruit. Hori- zontal axis labels represent variety by letter (K - Kordia, L- Lapin, R - Regina, Si - Simone, Sw - Sweetheart, Sy - Sylvia, V - Van) and season by number (1 - 2005/06, 2 -2006/07, 3 2007/08). A significant relationship between cracking index with total cracking incidence (P = 0.011) and incidence of side- cracked fruit (P<0.001) was found. No significant rela- tionship was found between cracking index and incidence of cuticular-cracked fruit. No differences were found be- tween values for stemless fruit and fruit with stems at- tached (both had index values of 25). 4. Discussion and Conclusions Lower crop loads resulted in a greater incidence of fruit cracking in sweet cherry. This was seen in both ma- nipulated crop loads and in the natural crop load survey. Responses to rainfall and the difference in magnitude of cracking between 2005/06 and 2006/07 confirm a strong seasonal impact on crack development, consistent with the findings of Measham et al. (2009). However, rainfall per se did not account for the differences in cracking between seasons, suggesting that other environmental parameters and fruit growth patterns are also important in the develop- ment of cracks. Fruit development is important in crack susceptibil- ity as all post bloom thinning showed increased cracking with lower crop loads. This implies that the effect of crop load on final cracking levels are determined post cell divi- sion, and cracking is therefore more likely to be attribut- able to cell expansion during the later stages of growth. Cell expansion is a function of internal water entry, which has been linked to increased rates of side-cracked fruit (Measham et al., 2010). This also supports the findings of Yamaguchi et al. (2002) who linked cracking suscepti- bility at harvest to cell length. Pre bloom thinning would therefore be the preferable option for manipulating crop load for size whilst minimising the risk of cracking. In addition, the development of the cuticle during early growth stages should be investigated with regard to cuticle integrity during the later periods of development. The number of cuticular-cracked fruits in low fruit load trees increased significantly in ‘Simone’. During cell expansion, relative canopy cover on a whole tree basis in low fruit load trees, compared to high load trees, may prevent moisture loss from the fruit surface through reduced airflow around fruit bunches in a timely and effective manner, confirming the importance of leaf:fruit ratio in quality management decisions (Whiting and Lang, 2004). In contrast to other studies (Spayd et al., 1986; Whiting and Lang, 2004) cracking susceptibility in this study did not seem to be related to fruit quality properties, nor did increased crop load limit fruit size or sugar accumulation, or enhance firmness (Christensen, 1996) in any of the ma- nipulated crop load trails. In the natural crop load survey, variation in fruit proper- ties was mostly influenced by variety, with crop load only further influencing fruit weight, but not size, sugar level, stem length or skin thickness. Cracking incidence was also not significantly correlated with the fruit properties record- ed. This is in contrast with accepted views that both fruit size (Simon, 2006) and sugar levels (Christensen, 1996) are closely linked with cracking. Studies suggesting veg- etative growth provides a strong photoassimilate sink in apricots (Costes et al., 2000) support these results where- by fruit crop load may not strongly influence source:sink relationships. Results from both the manipulated and natu- ral crop load trials do confirm studies that report no differ- ences in sugar levels between varieties of varying cracking susceptibility (Moing et al., 2004). 31 Furthermore, the results of the present study can be ex- plained by the level of crop load achieved under normal orchard practice. Fruit loads were relatively low in trees from this study with the majority being lower than 15 fruits per cm2 TCSA; the highest value reached was about 27 fruits per cm2 TCSA, or the equivalent of just over 2000 fruits on a tree with a trunk circumference of 30 cm. It is possible that fruit quality (size and sugars) was not di- minishing under this scenario as there were still available resources within the tree from which to draw. This finding highlights the strong potential for encouraging good fruit set, and subsequent crop load, as a practical and viable management tool in mitigating yield losses from rain-in- duced cherry fruit cracking, given the significant reduction in cracked fruit with increased crop loads. The incidence of cracking recorded in situ was correlat- ed with the cracking index for varieties in the natural crop load survey but for only one of the manipulated crop load trials. The cracking index procedure may not necessarily be reliable for predicting cracking susceptibility given the differences found in cracking incidence with crop load, with seasons (Measham et al., 2009), and when compared to other growing regions (Christensen, 1996; Greco et al., 2008). The strong correlation between cracking index and the incidence of side cracks recorded in situ supports the build up of turgor within the fruit as a likely driver of side cracking (Sekse, 1995), which can be somewhat mitigated by skin and cuticular properties. This could be due to dif- ferences in shape; curvature of the skin has been related to cracking susceptibility (Sawada, 1934), and could also explain why size was the only fruit parameter positively correlated with cuticular cracking. This study has confirmed that crop load management can be successfully used to mitigate cracking without compromising fruit size. The results from this study did not confirm the relationship between fruit size, or sugar, and the incidence or cracking, but highlight the impor- tance of skin properties in crack development. References AYALA M., LANG G.A., 2008 - C-13-photoassimilate parti- tioning in sweet cherry on dwarfing rootstocks during fruit development, pp. 625-632. - In: ERIS A., and M. BURAK (eds.) Proceedings of the V International Cherry Symposium. International Society for Horticultural Science, Vols. 1 and 2. CHRISTENSEN J.V., 1972 - Cracking in cherries. III. Determi- nation of cracking susceptibility. - Acta Agriculturae Scandi- navica, 22: 128-136. CHRISTENSEN J.V., 1996 - Rain-induced cracking of sweet cherries: Its causes and prevention, pp. 297-330. - In: WEBSTER A.D., and N.E. LOONEY (eds.) Cherries: Crop physiology, production and uses. CAB International, pp. 513. COSTES E., FOURNIER D., SALLES J.C., 2000 - Changes in primary and secondary growth as influenced by crop load in ‘Fantasme®’ apricot trees. - J. of Hortic. Sci. & Biotec., 75: 510-519. GRECO P., PALASCIANO M., MARIANI R., PACIFICO A., GODINI A., 2008 - Susceptibility to cracking of thirty sweet cherry cultivars. - Acta Horticulturae, 795: 379-382. HANSON E.J., PROEBSTING E.L., 1996 - Cherry nutrient requirements and water relations, pp. 243-258. - In: WEBSTER A.D., and N.E. LOONEY (eds.) Cherries: Crop physiology, production and uses. CAB International, Wallingford, UK, pp. 513. KAPPEL F., 1991 - Partitioning of above-ground dry matter in ‘Lambert’ sweet cherry trees with or without fruit. - J. Amer. Soc. for Hortic. Sci., 116: 201-205. LANG G.A., OLMSTEAD J.W., WHITING M.D., 2004 - Sweet cherry fruit distribution and leaf populations: Modeling can- opy dynamics and management strategies, pp. 591-599. - In: WEBSTER A.D. (ed.) Key processes in the growth and crop- ping of deciduous fruit and nut trees. International Society for Horticultural Science, pp. 715. MEASHAM P.F., BOUND S.A., GRACIE A.J., WILSON S.J., 2009 - Incidence and type of cracking in sweet cherry (Prunus avium L.) are affected by genotype and season. - Crop and Pasture Science, 60: 1002-1008. MEASHAM P.F., GRACIE A.J., WILSON S.J., BOUND S.A., 2010 - Vascular flow of water induces side cracking in sweet cherry (Prunus avium L.). - Adv. Hort. Sci., 24(4): 243-248. MOING A., RENAUD C., CHRISTMANN H., FOUILHAUX L., TAUZIN Y., ZANETTO A., GAUDILLERE M., LAIGRET F., CLAVERIE J., 2004 - Is there a relation between changes in osmolarity of cherry fruit flesh or skin and fruit cracking susceptibility? - J. Amer. Soc. for Hortic. Sci., 129: 635-641. PROEBSTING E.L., MILLS H.H., 1981 - Effects of season and crop load on maturity characteristics of ‘Bing’ cherry. - J. Amer. Soc. for Hortic. Sci., 106: 144-146. RICHARDSON D.G., 1998 - Rain-cracking of ‘Royal Ann’ sweet cherries: Fruit physiological relationships, water tem- perature, orchard treatments, and cracking index, pp. 677- 682. - In: YSTAAS J., O. CALLESEN (eds.) Third Interna- tional Cherry Symposium, Vols. 1 and 2. SAWADA E., 1934 - A physical consideration of the mechanism of the cracking of sweet cherries. - Transactions of the Sap- poro Natural History Society, XII: 365-376. SEKSE L., 1995 - Fruit cracking in sweet cherries (Prunus avi- um L.). Some physiological aspects - a mini review. - Scien- tia Horticulturae, 63: 135-141. SIMON G., 2006 - Review on rain induced fruit cracking of sweet cherries (Prunus avium L.), its causes and the possi- bilities of prevention. - International Journal of Horticultural Science, 12: 27-35. SPAYD S.E., PROEBSTING E.L., HAYRYNEN L.D., 1986 - Influence of crop load and maturity on quality and suscepti- bility to bruising of ‘Bing’ sweet cherries. - J. Amer. Soc. for Hortic. Sci., 111: 678-682. WHITING M.D., LANG G.A., 2004 - Effects of leaf area re- moval on sweet cherry vegetative growth and fruit quality, pp. 467-472. - In: WEBSTER A.D. (ed.) Key processes in the growth and cropping of deciduous fruit and nut trees. Inter- national Society for Horticultural Science, pp. 715. YAMAGUCHI M., SATO I., ISHAGURO M., 2002 - Influences of epidermal cell sizes and flesh firmness on cracking susceptibility in sweet cherry (Prunus avium L.) cultivars and selections. - J. of the Japanese Soci. for Hortic. Sci., 71: 738-746.