J 11:1 3.9 .. prfiuru :.n Link. a. 2.... ' >l1- .0 :1... , . 1.1:... 1.121.. I. .r. . . . .X. .. 3).. .. .1911 :.Hsn|l\htd~.5!;1 r . .v . vita. u . i. an 1.1.»."433 I». a if :11 31...... iv-fi. ‘ . . 2:34.31 , V , I . .4 w). 5" i . u x .3. .ylzap‘? . . . . . i: (t ‘ n. : . spun; Ilbl f’ .wrla. ) .quu‘: 3;; “55$: 3 :‘i .1: a .15.»! nr-Lw . TWIERSWY l l; l Illllljilll ll llllllllllllllllll 31 93 01415 0837 This is to certify that the thesis entitled CANOPY AND CROPPING INFLUENCE ON VINE GROWTH, PHYSIOLOGY, AND CLUSTER DISEASE INCIDENCE OF SEYVAL AND VIGNOLES GRAPEVINES. presented by Russell Paul Smithyman has been accepted towards fulfillment of the requirements for Masters degree in Horticulture <35 ' (”61/ / Major professor Date Aug. 5, 1994 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Mlchlgan State Unlverslty PLACE ll RETURN BOXto roman” mum your record. TO AVOID FINES rotum on or baton dd. din. DATE DUE DATE DUE DATE DUE CANOPY AND CROPPING INFLUENCE ON VINE GROWTH, PHYSIOLOGY, AND CLUSTER DISEASE INCIDENCE OF SEY VAL AND VIGNOLES GRAPEVINES By Russell Paul Smithyman A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Horticulture 1994 .I 3 i ‘N a“ r ' n. l J, 7w“ “*1 : ABSTRACT CANOPY AND CROPPING INFLUENCE ON VINE GROWTH, PHYSIOLOGY, AND CLUSTER DISEASE INCIDENCE OF SEY VAL AND VIGNOLES GRAPEVINES By Russell Paul Smithyman Seyval and Vi gnoles grapevines were subjected to decreasing pruning seventies to increase the level of fruitful (clusters) and vegetative (apical meristems) sinks during anthesis in an effort to reduce fruit-set, cluster compactness and Botrytis incidence. Fruit-set appears to be influenced more by cluster number than apieal meristems. Due to their large cluster size, Seyval showed a significant response to greater fruitful sink competition. However, the small clusters of Vi gnoles were insufficient in influencing f ruit-set at pruning seventies that produced sufficient vegetation for fruit and wood maturity. Increased shoot numbers per vine facilitated early filling of the trellis area with foliage. However, a limit existed where vegetative sink competition adversely affected light penetration into the eanopy, leaf production after veraison, and eane cold hardiness. Three eanopy I configurations (severely pruned, full trellis, and hedging), providing three levels of shoot production, were compared to examine their influence on shoot and leaf growth as well as leaf area formation. Both the full trellis and hedging systems increased early filling of the trellis area with foliage and yield, but canopy density beeame a problem in the hedged vines after veraison. A moderate pruning severity, accompanied by post-set thinning in Seyval appears to be the best method for creating a sufficient sink competition to reduce berry-set, cluster compactness, and thus, Botrytis infection. Dedieated to missed days in the woods and streams of Michigan. ACKNOWLEDGMENTS I would like to express my sincere appreciation to Dr. Stan Howell for his guidance, encouragement, and assistance during my tenure on this project. The opportunity to study viticulture under his tutelage has been enlightening and gratifying. I would like to thank the members of the guidance committee: Dr. E. Hanson and Dr. D. Dickman. I am deeply indebted to Doug Welsh for the use of his vineyard, and to his family for allowing me into their home during my trips to Penn Valley. Thanks are also due Dave Miller for his support as a coworker and friend. Lastly, I wish to thank my wife, Jenn, for helping me prepare this manuscript. Her patience and moral support during this endeavor was appreciated. TABLE OF CONTENTS Ease LIST OF TABLES ........................................ xii LIST OF FIGURES ........................................ x INTRODUCTION ......................................... 1 LITERATURE REVIEW ..................................... 4 Leaf Area ............................................ 4 Canopy Design ............. - ............................ 5 Carbohydrate Sinks ...................................... 6 Flower Initiation and Differentiation ........................... 7 Bloom and Fruit Set ..................................... 8 Berry Development ...................................... 9 Cultural Practices to Influence Fruit-Set ........................ 9 Chemical Influence on Fruit-Set ............................ 10 Cultural Control of Botrytis ............................... 11 Statement of Objectives .................................. 12 Literature Cited ........................................ 14 Chapter I. Influence of Canopy Configuration on Canopy Development, Yield, Fruit Composition, and Cane Quality of Seyval Grapevines. Abstract ............................................ 23 Introduction .......................................... 24 Materials and Methods ................................... 27 Results and Discussion .................................. 30 Conclusions .......................................... 35 Literature Cited ........................................ 37 Chapter II. Influence of Canopy Configuration on Shoot Growth, Leaf Development, and Leaf Area of Seyval Grapevines. Abstract ............................................ 54 Introduction .......................................... 54 Materials and Methods ................................... 58 Results and Discussion .................................. 61 Conclusions .......................................... 67 Literature Cited ........................................ 69 Chapter III. The Use of Carbohydrate Sink Competition to Reduce Fruit Set and Botrytis Rot in Seyval Grapevines. Abstract ............................................ 88 Introduction .......................................... 88 Materials and Methods ................................... 91 Results and Discussion .................................. 96 Conclusions .......................................... 99 Literature Cited ........................................ 101 Chapter IV. The Use of Carbohydrate Sink Competition to Reduce Fruit Set and Botrytis Rot in Vi gnoles Grapevines. Abstract ............................................ 118 Introduction .......................................... 1 18 Materials and Methods ................................... 120 Results and Discussion .................................. 124 Conclusions .......................................... 128 Literature Cited ........................................ 130 CONCLUSIONS ......................................... 150 Appendix 1. Influence of Cane Characteristics on Primary Bud and Periderm Cold Hardiness .............................. 153 vi LIST OF TABLES Table Page CHAPTER I 1. Influence of pruning treatments on vine yield from 1988 - 1993 of Seyval grapevines at Fenn Valley vineyards .......... 40 2. Influence of pruning treatments on node number at pruning and cluster number at harvest from 1988 - 1993 of Seyval grapevines at Fenn Valley vineyards .......................... 41 3. Influence of pruning treatments on cluster wt., berry wt., and berries/cluster at harvest from 1989 - 1993 of Seyval grapevines at Fenn Valley vineyards ................................. 42 4. Influence of pruning treatments on fruit composition and vine sugar production at harvest from 1989 - 1993 of Seyval grapevines at Fenn Valley vineyards .......................... 43 5. Influence of pruning treatments on incidence of Botrytis rot within and among clusters at harvest from 1989 - 1993 of Seyval grapevines at Fenn Valley vineyards ..................... 44 6. Influence of pruning treatments on vine size and growth per node retained at pruning from 1988 - 1993 in Seyval at Fenn Valley vineyards ................................. 45 7. Influence of pruning treatments on shoot population measured after verasion from 1991 - 1993 in Seyval at Fenn Valley vineyards ................................... 46 8. Influence of pruning g treatments on cane quality as assessed after leaf fall from 1991 - 1993 in Seyval at Fenn Valley vineyards ....................................... 47 9. Influence of pruning treatments on occurrence of shootless nodes to Seyval grapevines measured just past burst from 1991 - 1993 at Fenn Valley vineyards ......................... 48 vii to IQ CHAPTER II Influence of pruning system in 1993 on the yield components and the fruit quality indices of Seyval grapevines at Fenn Valley vineyards ................................... 72 Influence of pruning treatments in 1993 on the vine size indices of Seyval grapevines at Fenn Valley vineyards .............. 73 Influence of pruning system in 1993 on shoot leaf area at harvest of Seyval grapevines at Fenn Valley vineyards ............. 74 Influence of pruning system in 1993 on vine leaf area at harvest of seyval grapevines at Fenn Valley vineyards .............. 75 CHAPTER III Influence of pruning severity and time of cluster thinning on the vine size indices averaged from 1992 — 1993 of Seyval grapevines at Fenn Valley vineyards ..................... 103 Influence of pruning severity and time of cluster thinning on fruit set averaged from 1992 - 1993 of Seyval grapevines at Fenn Valley vineyards ................................. 104 Influence of pruning severity and time of cluster thinning on the vine size and yield indices averaged from 1992 - 1993 of Seyval grapevines at Fenn Valley vineyards ................... 105 Influence on increasing crop levels retained at thinning on yield (kg) in 1993 of Seyval grapevines atFenn Valley vineyards ............................. . . . 106 Influence of pruning severity and time of cluster thinning on the yield components and fruit quality indices averaged from 1992 - 1993 of Seyval grapevines at Fenn Valley vineyards .......... 107 CHAPTER IV Influence of the number of nodes retained and cane length on the vine size and shoot and cane characteristics of Vignoles grapevines at Fenn Valley vineyards in 1992 ............ 133 Influence of the number of nodes retained and eane length on the vine size and shoot and cane characteristics of Vignoles grapevines at Fenn Valley vineyards in 1993 ............ 134 viii Influence of the number of nodes retained and cane length on the yield and the components of yield of Vignoles grapevines at Fenn Valley vineyards in 1992 .............. 135 Influence of the number of nodes retained and eane length on the yield and components of yield of Vignoles grapevines at Fenn Valley vineyards in 1993 .............. 136 Influence of the number of nodes retained and cane length on fruit set and quality of Vi gnoles grapevines at Fenn Valley vineyards in 1992 .................... 137 Influence of the number of nodes retained and cane length on fuit set and quality of Vi gnoles grapevines at Fenn Valley vineyards in 1993 ............................. 138 LIST OF FIGURES Figure IQ CHAPTER I Effect of canopy configuration on seasonal canopy development averaged from 1991 - 1993 of Seyval grapevines at Fenn Valley vineyards, expressed as Page leaf layer number ...................................... 49 Frequency distribution averaged from 1991 - 1993 of the number of contacts (leaves and fruit) measured at bloom by point quadrat for Seyval grapevines pruned to three canopy configurations .............................. 50 Frequency distribution averaged for 1991 - 1993 of the number of contacts (leaves and fruit) measured at verasion by point quadrat for Seyval grapevines pruned to three canopy configurations .............................. 51 Frequency distribution averaged for 1991 — 1993 of the number of contacts (leaves and fruit) measured at harvest by point quadrat for Seyval grapevines pruned to three canopy configurations ..................... p ......... 52 CHAPTER II 1993 Standard for flowers per cluster measurements in Seyval grapevines .......................... 76 Standard for primary shoot leaf area measurement in Seyval grapevines .................................... 77 Standard for lateral shoot leaf area measurements in Seyval grapevines .................................... 78 Influence of canopy configuration on primary shoot length (cm) in Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 79 10. 11. Influence of canopy configuration on primary shoot leaf production in Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ..................... 80 Influence of canopy configuration on lateral shoot production in Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 81 Influence of canopy configuration on lateral shoot leaf production in Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 82 Influence of canopy configuration on lateral shoot growth (cm) in Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 83 Influence of canopy configuration on total shoot vegetative growth (cm) in Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ................ 84 Influence of canopy configuration on total leaves per shoot in Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 85 Influence of eanopy configuration on leaf area production in Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 86 CHAPTER III 1992 standard for flowers per cluster measurements in Seyval grapevines .................................... 108 1993 standard for flowers per cluster measurements in Seyval grapevines ...................................... 109 Effect of pruning severity on the 1992 season canopy development of Seyval grapevines at Fenn Valley vineyards, expressed as leaf layer number ............................. 110 Effect of pruning severity on the 1993 season canopy development of Seyval grapevines at Fenn Valley vineyards, expressed as leaf layer number of Seyval ...................... 1 1 1 xi Influence of pruning severity on the total vegetative length per shoot (primary + lateral) of Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ........... 112 Influence of pruning severity on the total leaves per shoot (primary + lateral) of Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ..................... 113 Influence of pruning severity on the lateral number per shoot (primary + lateral) of Seyval grapevines at Fenn Valley vineyards during the 1993 growing season ........... 114 1992 percentage Botrytis rot as a function of the number of benies set per cluster in Seyval grapevines .............. 115 1993 percentage Botrytis rot as a function of the number of benies set per cluster in Seyval grapevines .............. 116 CHAPTER IV 1992 standard for flowers per cluster measurements in Vignoles grapevines ................................... 139 1993 stande for flowers per cluster measurements in Vi gnoles grapevines ................................... 140 1991 and 1992 combined Vignoles primary bud hardness evaluation ..................................... 141 1991 and 1992 combined Vignoles peridenn hardness evaluation ..................................... 142 Influence of pruning severity on primary shoot length (cm) in Vignoles grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 143 Influence of pruning severity on primary shOot leaf production in Vignoles grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 144 Influence of pruning severity on lateral shoot production in Vignoles grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 145 xii 10. 11. Influence of pruning severity on lateral leaf production in Vignoles grapevines at Fenn Valley vineyards during the 1993 growing season ............................ 146 Influence of pruning severity on lateral shoot length (cm) in Vignoles grapevines at Fenn Valley vineyards during the 1993 growing season ............................ .147 Influence of pruning severity on total vegetative length per shoot (primary + lateral) in Vi gnoles grapevines at Fenn Valley vineyards during the 1993 growing season ................ 148 Influence of pruning severity on total leaves per shoot (primary + lateral) in Vi gnoles grapevines at Fenn Valley vineyards during the 1993 growing season ................ 149 xiii !J 10. 11. APPENDIX Influence of cane diameter on the primary bud cold hardiness of Seyval grapevines ........................ 155 Influence of cane diameter on the periderm cold hardiness of Seyval grapevines ........................ 156 Influence of intemode length on the primary bud cold hardiness of Seyval grapevines ........................ 157 Influence of intemode length on the periderm cold hardiness of Seyval grapevines ........................ 158 Influence of periderm color on the primary bud cold hardiness of Seyval grapevines ........................ 159 Influence of periderm color on the periderm cold hardiness of Seyval grapevines ........................ 160 Influence of lateral status on the primary bud cold hardiness of Seyval grapevines ........................ 161 Influence of lateral status on the periderm cold hardiness of Seyval grapevines ........................ 162 Influence of node position along the cane on the primary bud cold hardiness of Seyval grapevines ............. 163 Influence of node position along the cane on the periderm cold hardiness of Seyval grapevines ............... 164 Influence of cane characteristics on primary bud cold hardiness of Seyval grapevines at Fenn Valley vineyards .......... 165 Influence of eane characteristics on periderm cold hardiness of Seyval grapevines at Fenn Valley vineyards .......... 166 xiv Introduction VineyardManagememfloals The viticulturalist’s objective is to manage each grapevine to produce the greatest consistent yield of high-quality fruit, while maintaining vine health. Cultural management research has increased the understanding of canopy configuration, structure, and function. However, few practices have been adopted commercially by growers because of differences in cultivars, climates, and soils. An understanding of specific circumstances of cultivar growth, localized environment and costs of production are required if the viticulturalist’s objective is to be achieved. Cmpflontmlandlargeflusteredflnltivars Many phylloxera resistant grape cultivars are economically important in Michigan viticulture. However, because their growth and fruiting habits often differ from the widely cultivated Vitis labruscana Bailey grapevines, dormant-season balanced pruning, a proven method of controlling crop level and maintaining fruit quality of V. labruscana (11,44), can fail to provide adequate crop control for many large clustered hybrid cultivars (22,36,37,46). Some cultivars can produce three or four heavy (>500 gm) clusters per shoot as well as fruitful lateral shoots (59). Fruitful shoots can also arise from base buds that are not counted in the balanced pruning practice (46,57). For such high producing cultivars, increasing the pruning severity is insufficient to control crop load and maximize filling of the trellis space with leaves early in the growing season (46,57). Overproduction is detrimental to vegetative growth and fruit maturity (11,22,25,26,31,44,46,61,64, 90,96), but can be corrected by cluster thinning (22,26,36,37, 61,96) so that, a balance 1 2 between vegetative growth and fruiting is produced. However, thinning increases labor and production expense for such cultivars. Severe pruning is presently employed to limit cluster thinning costs. Such pruning delays canopy formation early in the growing season; maximum leaf area for the trellis space is not reached until well after bloom. The potential loss of photosynthesis due to the delay in canopy formation is of concern. A more cost- effective method of crop control and canopy management that would better utilize vine capacity would be very useful. (1118th Even with successful cropping procedures, pathogens can upset the balance achieved between vegetation and fruit, causing damage to vine health and fnrit quality. Cultivars differ in their vulnerability to different pathogens. Seyval has a serious harvest season cluster rot problem due primarily to Botrytis cinerea Pers. The most serious infections occur post-veraison during fruit ripening after rains, when the foliage layers are fully developed and drying of the clusters is inhibited. Cluster exposure and berry contact within the cluster influence Botrytis rot incidence and intensity of infection (54). At present, an extensive chemical spray schedule is needed to prevent infection. However, many important fungicides are being banned or their use limited. Cultural practices are needed that would reduce cluster rot infection and decrease dependance on the use of chemicals normally applied for control. The cultivars Seyval (S.V. 5276) and Vignoles (Ravat 51) were chosen for this study because both are of importance in Michigan’s wine industry and both have serious cluster rot problems. In addition, Seyval is easily overcropped because it forms large, compact clusters. Seyval is managed using severe balanced pruning and flower cluster thinning (37). It is recommended that a 15+10 balanced pruning formula with thinning to 1.5 clusters per shoot be employed to control crop level (61). 3 Vignoles is by contrast, a small-clustered cultivar that also bears compact clusters susceptible to Botrytis infection. Because of the smaller clusters, balanced pruning alone is sufficient for crop control. A 15+15 pruning formula on a High Cordon training system is recommended (28). More practical cultural methods need to be investigated in the production of both these cultivars. Importantly, information learned on these two should provide a basis for similar problems in other cultivars such as Chancellor and Pinots noir, gris and blanc. Literature Review meArea The photosynthate produced during the vegetative season must drive the processes of vegetative growth, fruiting, and flower bud development for next years crop. Reserves must also be stored over the winter for cold hardiness, and initiation of growth and development, until a new canopy is producing photosynthate the following spring. Equally as signifieant is the mobilization and transport of assimilate between different organs of the vine in order to carry out these processes. Leaf development is important for fruit and shoot maturity. A leaf number greater than 10 per shoot is necessary for optimal fruit ripening and shoot weight (87,93). Leaf photosynthesis is greatest when the leaf is fully expanded, and then gradually declines shortly thereafter (56). The amount of photosynthate produced by each leaf is important for the development of the subtending bud (24,79). Over 80 percent of the assimilate a cluster collects is produced by the leaves on the same side of the shoot (47,48). Buttrose (8) found that as leaf number was reduced, root dry weight decreased most severely, followed by reductions in berry development, and then trunk and shoot carbohydrate reserves. Increasing the number of leaves per shoot increases berry sugar, weight, and coloration at harvest (87). However, the production of each new leaf adds to the vine leaf area and the formation of leaf layers within the canopy. The leaf area of the vine is critical because it produces the assimilate necessary for crop ripening and wood maturity. A minimum leaf area of 7-14 cmA2 per gram of fruit is needed for adequate fruit ripening (31,83). Smart showed that a dense canopy can reduce light levels to 1% of the solar radiation in the fruiting and renewal zone (80), and an optimal eanopy has no more than three leaf layers (81). Light penetration into the canopy is important for photosynthesis (35,78), f ruitf ulness (27,74), flower bud formation (74,80), 5 fruit set (15,68), fruit quality (63,65,82), yields (74), and wood hardiness (25,74,85). Thus, it is evident that a maximum leaf area exists where light penetration into the canopy is restricted to the point that vine productivity and conditions begin to diminish. The maintenance of this leaf area is important to maximize the amount of assimilate produced in the vegetative season for vine growth and development. When this maximum leaf area to facilitate vine grth is established is of great interest. l4C-labeled assimilates move from the leaves to the perennial portions of the vine in autumn, and then reappear in the shoot growth the following spring (73). This new growth is dependent on the remobilization of stored carbohydrates from the previous year (98,99). Photosynthesis increases up to the fourth week after anthesis and then declines until harvest (53). Cluster development is slow until the leaf area is established (95). Thus, it is important to fill the trellis area with vegetative growth as early as possible to facilitate vine growth and development without creating a dense canopy later in the season. Canopyfleaign Manual labor for pruning and thinning has become increasingly scarce and costly. Much attention has been focussed on minimal pruning management techniques to reduce labor. Mechanical hedging, selective hand pruning, or no pruning, increase the number of buds retained for potential shoot production from the previous years vegetative growth compared to conventional pruning. Downton and Grant (20) showed that an increase in shoot number resulted in a change in vine morphology; smaller shoots were produced compared to spur pruned vines. Minimally pruned vines had a greater leaf area and higher photosynthetic rates before bloom, but little leaf production after, as spur pruned vines had a greater leaf area at harvest. The total amount of carbon fixed per vine was similar between treatments, but proportioned differently. Minimally pruned vines allocated more photosynthate to fruit ripening than to wood maturity. They concluded that minimally 6 pruned vines lose less of their fixed carbon from the previous year due to pruning. This was supported by Ruhl and Clingeleffer (69), who found that minimally pruned vines stored more carbohydrates in old wood but less in their canes and roots. Without any other crop control, minimal pruning increases yield as a function of increased cluster number, but brix are usually lower (41). Increasing shoot number increases both canopy density (81) and the labor expense incurred by cluster thinning fruitful varieties. Mechanical hedging has been suggested as an alternative to hand, balanced pruning, since it may increase labor efficiency three to six-fold (75,83). Continuous use of hedging requires the vineyard to be uniform in vine size and vigor (45) and without eane selection or shoot positioning, it has been shown to be useful for only a few years (45,55,66). It increases the amount of vegetation per foot of trellis space (43,45,55), allowing for earlier filling of the canopy. However, mechanical pruning for more than one to three years results in reduced yields, fruitfulness and fruit quality (43,45,55). Mechanical hedging may be a viable alternative when it is done in conjunction with selective hand pruning techniques (66). CarlnlrydrateSinks The organs competing for photoassimilates are termed sinks and the control of assimilate transport is determined by the relative position of the source leaves and sink organs, and the relative strength of the sink organs to accumulate assimilates (48). From bud-burst to bloom, the rapidly expanding leaves and shoot apex are the primary sinks for stored carbohydrates (95). Flower clusters compete for carbohydrates during bloom (60). After anthesis, the shoot apex and the developing clusters become the strongest sinks (15,24,60). At harvest, the vine storage sinks compete for assimilate. Only by balancing the source to sink ratio, can productive grapevines be maintained. An excessive crop creates a situation where the total fruiting sink is greater than any Ull ll mot clux' Hilc‘l F” i WES. 7 other sink. In response to excess crop, vine size decreases as shoot growth diminishes (22,25,26,61,64). Overcropping negatively effects fruit quality, beny weight, cluster weight, and berries per cluster (26,31,62). Fruit and wood maturity are delayed (2,5,61,92,97). The increased competition from high crop levels can influence morphological characteristics such as a decrease in shoot length, leaf size, cluster size, cluster weight, and berry weight (4,30,90). These canopy reductions improve canopy microclimate (63,64) and increase photosynthetic efficiency through reduced shading (83). Lateral shoots supply assimilate to clusters on the main shoot, but when fruitful, the mid- season clusters they produce draw assimilate from the main shoot (34). In total, an excessive crop level disrupts the source to sink balance, and is detrimental to vine growth and health. El 1.. . IE’EI .. Each node formed on a grapevine shoot contains three partially developed shoots with leaf and flower primordia for the following year (95). The apical buds formed late in the growing season are the least mature along the cane in the fall. Because the basal buds of a shoot are close in proximity to the storage and fruiting sinks, they accumulate carbohydrates later than the more apieal nodes, and are less fruitful in most cultivars (95). The French-American hybrids have fruitful basal buds (46,57). The leaf subtending each developing bud is the main source of assimilate for that bud (24,79). Thus, the light intensity on that leaf and bud is important during flower bud development (40,79), and Shading can decrease vine f nritf ulness (27,74). Flower initiation begins in early summer, when each bud begins to swell (58,84). BUds mature throughout the vegetative season. Differentiation from vegetative to fruiting Plimordia occurs in late summer until the vine acclimates to winter. The calyx, corolla, Slamws, and pistil are differentiated the following spring (72). Two carpels makeup the 8 pistil with two ovules in each carpel. With development of the stamens and differentiation of the pollen, the perfect inflorescence are ready to bloom. BloonLandEmitzset The basal clusters are the first to bloom, six to eight weeks after bud burst. Blooming lasts one to three weeks, with temperature being the controlling factor (95). High relative humidity and rain hamper pollination, while moisture stress decreases fruit set for up to four weeks following bloom (1). Grape flowers are either self-pollinated or aided by insects and wind (95). Fruit set is best at temperatures between 20 and 30 C (9,86). Two theories exist as to what regulates fruit set in grape vines. One suggests that growth hormones are transported to the cluster to induce set (88,89), while the other proposes that carbohydrate supply determines fruit set (49). Inna-vine competition for carbohydrates and growth hormones by vegetative (90) and fruiting sinks ( 17) have been shown to have a negative effect on fruit set Reducing sink competition and growth hormone sources by removing shoot tips, increases fruit set (15,76). Decreasing assimilate sources by leaf removal during bloom decreases fruit set (10). Thinning flower- clusters increases pollen genuinability and fruit set (94). Within-cluster competition decreases fruit set (37). Mullins (49) demonstrated that fruit set can occur in flowers cultured in vitro on a medium of sucrose and nutrients, suggesting that fruit set is regulated by assimilate supply. Roper et. al. (67) suggest that competition for photosynthate may be the limiting factor in cranberry fruit set. Strawberry fruit set and initial growth is not limited by the ability of receptacles to mobilize current source leaf assimilates; and early fruit grth is not related to pool sizes ( l9). Nitsh (51) removed the achenes of strawberries and the receptacle ceased to enlarge. However, if the berry was coated with a lanolin paste containing auxin, it enlarged normally. It is assumed the strawberry receptacle requires auxin for normal growth (52). barn Cmm l'ariet full cc stems] and ne H en‘r Emily; Bfifld Crumb; 9 In grapevines, a direct correlation is reported between berry size and seeds per berry (12). Seed number has little relationship with auxin or abscisic acid content (9). Coombe (14) found increased auxin levels and improved fruit set in girdled vines. Varieties that undergo normal set contain higher auxin levels and greater berry size with a full complement of developing seeds than varieties that are stimulative parthenocarpic or stenospermocarpic (95). A competition for carbohydrates and growth hormones exists, and needs further experimentation to determine optimum cultural practices. BerrsLDeyelornnem When the benies become active sinks, many of their tissues become meristematic and growth begins. Those berries that have not initiated this process generally abscise. Berry development follows a double sigmoid growth curve consisting of three stages (18). Coombe (18) suggests that cell division before verasion and cell expansion after anthesis are the major determinants of the increase in weight of fleshy fruits. Some contributions are made by cell division after anthesis and an increase in concentration of solutes. Little is known about the influences upon early cell division and its differences may commonly contribute to fruit size variation. Early availability of carbohydrates and growth hormones would appear to be of great importance during this time. Manipulating the carbohydrate source/sink balance, can alter fruit set in grapevines. Shoot tipping at bloom (13,15,76), girdling (5,13), flower-cluster thinning (91,92), and the application of growth regulators (5,13,16) increase fruit set. Heavy pruning (91) or leaf removal during bloom decreases fruit set (10). Increasing the light intensity for photosynthesis promotes fruit set (68). It follows that vine training to allow more light reception into the canopy could increase fruit set (‘77). 1O Pruning is presently the most cost effective way to control crop level. Balanced pruning uses the one-year-old cane prunings as an estimate of vine size to determine the vines capacity for growth and production (3). Capacity is defined as the vines ability for total vegetative and reproductive growth (95). Vegetative growth can then be balanced with expected crop level to maintain vine size and sufficiently ripen the fruit (26). As stated earlier, easily overcropped cultivars, such as Seyval, require thinning. Thinning increases the ratio of leaves to fruit, improving the nutrition of both the vine and fruit (95). Some cultivars are flower-cluster thinned (37). Flower-cluster thinning increases pollen gerrninability, fruit set (94), and shoot growth (22,37). Post-set thinning decreases fruit set by increasing the fruiting and vegetative sink competition for assimilate and growth hormone (17,90), as well as reducing pollen quality. Reduced fruit set decreases cluster compactness. Post-set thinning reduces berries per cluster in Seyval, and a compensatory increase in berry size and fruit quality follows (29). Vegetative vigor and the development of laterals decreases as thinning is delayed until after fruit set (29). The period for source/sink manipulation of fruit set is suggested to extend from bloom to two weeks post-set (10). fllemicallnfluencemEnukSet The use of growth regulators is an altemative to hand thinning. Both ethephon and gibberellic acid have been used as berry abscission agents (36,37,91). Ethephon’s effectiveness is dependant upon environmental conditions, time of application, concentration, and cultivar (91). Gibberellic acid response has not been consistent (36,37). Chlormequat (CCC) application increases fruit set under good conditions (5,16) by diverting organic nutrients from the shoot tips to the developing ovaries (60). More partially seeded berries are set (5). Roubelakis and Kliewer (68) were unable to induce fruit set with CCC applications during poor light and temperature regimes. 11 Culturalfiomrolofiflatgztis The control of Botrytis cinerea Pers. is a costly endeavor for the vineyard manager. Cluster infection usually occurs late in the growing season, decreasing yield and fruit quality. B. cinerea survives the winter by forming sclerotia either on the surface or within plant tissue. Clusters and dead wood retained in the vineyard from the previous season can be an important source for inoculum in the spring the following year. The fungus lives most of its life as a saprophyte, getting nutrients from dead or dying plant parts (39). It may infect grape flowers and remain latent in these tissues until the clusters begin to mature (42). The spores of B. cinerea require prolonged periods of free water and nutrients on the berry surface for germination (39,50). Enzymes produced by the fungus can destroy the integrity of the berry in less than 24 hours (39). The grape berry’s main protection to Botrytis infection is its skin. The cuticle membrane and the epicuticular wax, provide the primary physical barrier to pathogen invasion (2). For berry infection to occur, the pathogen must find a weakness on the berry surface where it can bypass the cuticle membrane, or directly penetrate the surface (6). The epicuticular wax layer influences the retention of pesticides, the water retention of the berry surface, and the adhesive ability of plant pathogens (2). Percival et al. (54) found that berries of exposed clusters and those having little berry contact, produced more epicuticular wax and cuticle. Infection can be reduced by improving the microclimate within the canopy (39,70,71). By reducing the periods of free water within the canopy and increasing cluster exposure through canopy management, disease pressure within the vineyard can be reduced (39). leaf removal in the fruiting zone decreases infection (21,23,33,54). Since the leaf area is the source for photoassimilate, defoliation can have a detrimental affect on fruit quality (10,25,38), vine growth (7,32), and hardiness (25,38,85). Grapevines are 1") able to compensate for the loss in leaf area by producing lateral shoots and increasing their photosynthetic efficiency (7,10,32). However, fruitful laterals draw assimilate from the main shoot (34). Defoliation in moderation can improve the canopy microclimate at the expense of assimilate production. The canopy microclimate can be improved to decrease Botrytis cinerea incidence, through better design (70,71,81). Statement of Objectives An efficient canopy maintains enough leaf area to produce quality fruit and a sufficient number of mature buds. It allows for adequate light penetration into the fruiting zone of the canopy to ripen fruit and inhibit Botrytis infection. Some cultivars are more susceptible to cluster rot because they bear compact clusters. This cluster structure holds moisture within, facilitating infection. More economical practices are needed to address this problem. Seyval and Vignoles are valuable cultivars for white wine production in the state of Michigan. However, their growth habits pose some problems in producing quality fruit. Seyval produces large, compact clusters that have a tendency to contract Botrytis at harvest. Presently it is recommended that Seyval be severely pruned and cluster thinned to control overcropping. This inhibits early filling of the trellis area. Vignoles is a small clustered cultivar that also has a tendency to contract Botrytis rot at harvest The removal of reproductive and vegetative sinks diverts photoassimilate to retained sinks. The time of cluster and whole shoot removal influences assimilate supply per alternate sink during the critical period of fruit set. 13 II 1.. [I' I, . H L) or To investigate economical pruning techniques that will allow for expedient filling of the trellis for optimum fruit ripening in Seyval. To explore the effect of post-set cluster thinning to find an optimum method for decreasing Botrytis rot and improving fruit quality in Seyval. To examine the effect of post-set pruning on fruit set and Borryris infection in Vi gnoles. II I“! l' I l [1' I . . Hedging is a viable practice for pruning Seyval grapevines and maintaining quality yields and vine health. A canopy constructed of optimally spaced shoots will have superior architecture for fruit and wood maturity. . Increasing the amount of fruiting sinks at fruit set will decrease the number of berries SCI. Increasing the number of vegetative sinks at fruit set will decrease the number of benies set. Decreasing the number of benies set by post-set thinning and pruning will decreases the amount of Bonytis rot. The loss in yield in Seyval because of the decrease in fruit set can be overcome by retaining more clusters per vine during thinning. By using the competition among fruiting and vegetative sinks, I believe that more economical and environmentally sound canopy management practices can be adopted to improve fruit quality and health of the vine. to Literature Cited . Alexander, D. 1965. The effect of high temperature regimes or short period of water stress on development of small fruiting Sultana vines. Austral. J. A gr. Res. 16:817-823. Baker, EL. 1982. Chemistry and morphology of plant epicuticular waxes. In: Cutler, D.F., Alvin K.L., Price, C.E. (Eds): The Plant Cuticle, 139-161. Acedemic Press, Toronto. Bell, H.K., R.P. Larson and A.L. Kenworthy. 1958. Balanced pruning of the Concord grape - another look under present-day conditions. Quarterly Bull. Mich. Agr. Exp. Station. Michigan State Univ. 40:915-919. Bravdo, B., Y. Hepner, C. Loinger, S. Cohen, and H. Tabacman. 1984. Effect of crop level in a high yielding Cari gnane vineyard on growth, yield, and wine quality. Am. J. Enol. Vitic. 35:247-252. Brown, K., D.I. Jackson, and GE Steans. 1988. Effects of Chlormequat, girdling, and tipping on beny set in Vitis vinifera L. Am. J. Enol. Vitic. 39(1):91-94. Bulit, J., and D. Dubos. 1988. Botrytis bunch rot and blight. In: Pearson, R.C., Goheen, A.C. (Eds): Compendium of grape diseases, 13-15. APS Press, St. Paul. Buttrose, MS. 1966. The effect of reducing leaf area on the growth of roots, stems, and berries of Gordo grapevines. Vitis. 5:455-464. Buttrose, MS. 1968. The effect of varying berry number on Gordo grapevines with constant leaf area. Vitis 7:299-302. Buttrose, MS. and CR. Hale. 1973. Effect of temperature on development of the grapevine inflorescence after budburst. Am. J. Enol. Vit. 24: 14—16. 10. Candolfi-Vasconcelos, MC. and W. Koblet. 1990. Yield, fruit quality, bud fertility and starch reserves of the wood as a function of leaf removal in Vitis vinifera - Evidence of compensation and stress recovering. Vitis. 29:199-221. 11. Cawthon,D.L. and JR. Morris. 1977. Yield and quality of Concord grapes as affected by pruning severity, nodes per bearing unit, training system, shoot positioning, and sampling date in Arkansas. J. Amer. Soc. Hort. Sci. 102:760-767. 12. Cawthon, D.L. and JR. Morris. 1982. Relationship of seed number and maturity to berry development, fruit maturation, hormonal changes, and uneven ripening of Concord grapes. J. Amer. Soc. Hort. Sci. lO7(6):1097-1104. 14 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 24. 25. 26. 15 Coombe, B.G. 1959. Fruit set and development in seeded grape varieties as affected by defoliation, topping, girdling, and other treatments. Am. J. Enol. Vitic. 10:85-100. Coombe, B.G. 1960. Relationship of growth and development to changes in sugar, auxin, and gibberellins in fruit of seeded and seedless varieties of Vitis vinifera. Plant Physiol. 35:241-250. Coombe, B.G. 1962. The effect of removing leaves, flowers and shoot tips on fruit set of Vitis vinifera L. J. Hort. Sci. 37: 1-15. Coombe, B.G. 1970. Fruit set in grapevines: The mechanism of the CCC effect. J. Hort. Sci. 45:415-425. Coombe, B.G. 1973. Regulation of set and the development of the grape beny. Acta. Horticulture 34:261-273. Coombe, B.G. 1976. The development of fleshy fruits. Ann. Rev. Plant Physiol. 27(2):507-528. Darnell, R.L. and CC. Martin. 1988. Role of assimilate translocation and Carbohydrate accumulation in fruit set of strawberry. J. Amer. Hort. Sci. 113(1): 114-1 18. Downton, W.J.S. and W.J.R. Grant. 1992. Photosynthetic physiology of spur pruned and minimal pruned grapevines. Aust. J. Plant Physiol. 19:309-316. English, J.T., C.S. Thomas, J.J. Marois, and W.D. Gubler. 1989. Microclimates of grapevine canopies associated with leaf removal and control of Botrytis bunch rot. Phytopathology. 79(4):395-40l. Fisher, K.H., O.A. Bradt, J. Wiebe and V.A. Dirks. 1977. Cluster thinning ‘de Chaunac’ French hybrid grapes improves vine vigor and fruit quality in Ontario. J. Amer. Soc. Hort. Sci. 102(2):l62-165. Gubler, W.D., J.J. Marois, A.M. Bledsoe, L.J. Bettiga. 1987. Control of Botrytis bunch rot of grape with canopy management. Plant. Dis. 71(7):599-601. Hale, CR, and PJ. Weaver. 1962. The effect of developmental stage on direction of translocation of photosynthate in Vitis vinifera . Hilgardia 33:89-131. Howell, G.S., B.S. Stergios and SS. Stackhouse. 1978. Interrelation of productivity and cold hardiness of Concord grapevines. Am. J. Enol. Vitic. 29: 187-191. Howell, G.S., T.K. Mansfield and J.A. Wolpert. 1987. Influence of training system, pruning severity, and thinning on yield, vine size and fruit quality of Vidal blanc grapevines. Am. J. Enol. Vitic. 38(2): 105-1 12. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 16 Howell, G.S., D.P. Miller, C.E. Edson and R.K. Streigler. 1991. Influence of training system and pruning severity on yield, vine size, and fruit maturity of Vignoles grapevines. Amer. J. Enol. Vitic. 42(3):]91-198. Howell, G.S. May 1992. Pruning and training Vignoles grapevines in Michigan. Vintner & Vineyard. Coop. Ext. Serv. Pub. 6(1):1-6. Hunter, D.M. and J.T.A. Proctor. 1985. Effects of timing of thinning influence on Seyval blanc fruit yields and quality. Presented at 10th Annual meeting of the American Society for Enology and Viticulture/Eastern section, Mississauga, Ont. (August 1985). Kaps, ML. and GA. Cahoon. 1989. Berry thinning and cluster thinning influence on vegetative growth, yield, fruit composition, and net photosynthesis of Seyval blanc grapevines. J. Amer. Soc. Hort. Sci. 114(1):20-24. Kliewer, W.M. and RJ. Weaver. 1971. Effect of crop level and leaf area on growth, composition and correlation of ‘Tokay’ grapes. Am. J. Enol. Vitic. 23: 172-177. Kliewer, W. M and R. D. Fuller. 1973. Effect of time and severity of defoliation on growth of roots, trunk and shoots of Thompson Seedless grapevines. Am. J. Enol. Vit. 24: 59-64. Kliewer, W.M. and A.M. Bledsoe. 1987. Influence of hedging and leaf removal on canopy microclimate, grape composition, and wine quality under Califomia conditions. Acta. Hort. 206: 157-168. Koblet, W. 1977. Translocation of photosynthate in grapevines. In: Proc. OIV symp. on the quality of the vintage. pp.45-51. OVRI, Stellenbosch, South Africa. Kriedemann, P.E., B.R. Loveys, and W.J.S. Downton. 1975. Internal control of stomatal physiology and photosynthesis. 2. Photosynthetic responses to phaseic acid. Aust. J. Plant Physiol. 2:553-567. Looney, NE. and DP. Wood. 1977. Some cluster thinning and gibberellic acid effects on fruit set, berry size, vine growth, and yield of DeChaunac grapes. Canadian J. Plant Sci. 57:653-659. Looney, NE 1981. Some growth regulator and cluster thinning effects on berry set and size, berry quality, and annual productivity of ‘De Chaunac’ grapes. Vitis 20: 22-35. Mansfield, T.K. and GS. Howell. 1981. Response of soluble solids accumulation, fruitfulness, cold resistance, and the onset of bud growth to different defoliation stress at verasion in Concord grapevines. Am. J. Enol. Vitic. 32:200—205. Marios, J.J. 1987. Botrytis bunch rot of grapes. Oregon Dpt of Agr.: The Wine Advisory Board Research Report June (S):1-6. 43. 45. 47. 49. 51. 52. 53. 17 May, P. and A.J. Antcliff. 1963. The effect of shading on f ruitfulness and yield of the Sultana J. Hort. Sci. 38:85-94. . McCarthy, M.G., and RM. Cirami. 1990. Minimal pruning effects on the performance of selections of four Vitis vinifera cultivars. Vitis 29:85-96. . McClellan, W.D. and W.B. Hewitt. 1973. Early Botrytr's rot of grapes: Time of infection and latency of Botrytis cinerea Pers. in Vitis vinifera L. Phytopathology. 63: 1 151-1 157. Morris, J.R., D.L. Cawthon, and J.W. Flemming. 1975. Effect of mechanical pruning on yield and quality of Concord grapevines. Ark. Farm Res. pub. 24(3):1-12. Morris, JR. and D.L. Cawthon. 1980. Yield and quality responses of Concord grapes to training systems and pruning severity in Arkansas. J. Amer. Soc. Hort. Sci. 105:307-310. Morris, JR. and D.L. Cawthon. 1981. Yield and quality responses of Concord grapes to mechanized pruning. Am. J. Enol. Vitic. 32(4):280-282. Morris, J.R., C.A. Sims, J.E. Bourque and J.L. Oakes. 1984. Influence of training system, pruning severity and spur length on yield and quality of six French- American hybrid grape cultivars. Am. J. Enol. Vitic. 35:23-27. Motomura, Y. 1990. Distribution of l4C-assimilates from individual leaves on clusters in grape shoots. Am. J. Enol. Vitic. 41:306-312. Motomura, Y. 1993. 14C-assimilate partitioning in grapevine shoots: Effects of shoot pinching, girdling of shoot, and leaf -halving on assimilate partitioning from leaves into clusters. Am. J. Enol. Vitic. 44: 1-7. Mullins, M. G. 1967. Regulation of fruit set in the grapevine. Aust. J. Biol. Sci. 20: 1 141-1 147. Nelson, KB. 1951. Factors influencing the infection of table grapes by Botrytis cinerea (Pers). Phytopathology. 41:319-326. Nitsch, JP. 1950. Growth and morphogenesis of the strawberry as related to auxin. Amer. J. Bot. 37:211-215. Nitsch, JP. 1953. The physiology of fruit growth. Ann. Rev. Plant Physiol. 4:199-236. Pandey, RM. and H.L. Farmahan. 1977. Changes in the rate of photosynthesis and respiration in leaves and berries of Vitis vinifera grapevines at various stages of berry development. Vitis. 16:106-111. 55. 57. 59. 61. 65. I8 Percival, D.C., J.A. Sullivan, and K.H. Fisher. 1993. Effect of cluster exposure, berry contact and cultivar on cutical membrane f orrnation and occurrence of bunch rot (Botrytis cinerea Pers.:Fr.) with 3 Vitis vinifera L. cultivars. Vitis 32:87-97. Pollock, J.G., ES. Shepardson, NJ. Shaulis, and DE. Crowe. 1977. Mechanical pruning of American hybrid grapevines. Trans. ASAE 20(5):817-821. Poni, S., C. Intrieri, and O. Silvestroni. 1994. Interactions of leaf age, fruiting and exogenous cytokinins in Sangiovese grapevines under non-ini gated conditions. 1. Gas exchange. Am. J. Enol. Vitic. 45(1):71-78. Pool, R.M., C. Pratt, and H.D. Hubbard. 1978. Structure of base buds in relation to yield of grapes. Amer. J. Enol. Vit. 29:36-41. Pratt, C. 1971. Reproductive anatomy of grapes - A review. Amer. J. Enol. Vitic. 22292-109. Pratt, C. 1974. Vegetative anatomy of cultivated grapes- -A review. Amer. J. Enol. Vit. 25:131- 150. Quinlan, JD. and R.J. Weaver. 1970. Modification of pattern of the photosynthate movement within and between shoots of Vitis vinifera L. Plant Physiol. 46: 527-530. Reynolds, A.G., R.M. Pool and LR. Mattick. 1986a. Effect of shoot density and crop control on growth, yield, fruit composition and wine quality of ‘Seyval blanc’ grapevines. J. Amer. Soc. Hort. Sci. 111(1):55-63. . Reynolds, A.G., R.M. Pool andL.R. Mattick. 1986b. Influence of cluster exposure on fruit composition and wine quality of ‘Seyval blanc’ grapes. Vitis 25:85-95. . Reynolds, A.G., and DA. Wardle. 1989a. Impact of various canopy management techniques on growth, yield, fruit composition and wine quality of Gewurtztraminer. Am. J. Enol. Vitic. 40:121-129. Reynolds, AG. and DA. Wardle. 1989b. Effects of timing and severity of summer hedging on growth, yield, fruit composition, and canopy characteristics of ‘De Chaunac’ 1. Canopy characteristics and growth parameters. Am. J. Enol. Vitic. 40:109-120. Reynolds, AG. and DA. Wardle. 1989c. Effects of timing and severity of summer hedging on growth, yield,fruit composition, and canopy characteristics of ‘De Chaunac’ 2. Yield and fruit composition. Am. J. Enol. Vitic. 40:299-308. Reynolds, AG. and DA. Wardle. 1993. Yield component path analysis of Okaagan Reisling vines conventionally pruned or subjected to simulated mechanical pruning. Am. J. Enol. Vitic. 44(2):]73-179. 67. 69. 70. 71. 72. 74. 75. 76. 79. 19 Roper, T.R., EJ. Stang, and GM. Hawker. 1992. Early season leaf removal reduces fruit set and size in cranberry (Vaccinium macrocarpon Ait.). Hort. Science. 27(1):75. Roublelakis, K.A. and W.M. Kliewer. 1976. Influence of light intensity and growth regulators on fruit set and ovule fertilization in grape cultivars under low temprature conditions. Am. J. Enol. Vitic. 27(4):163-167. Ruhl,E.H. and PR. Clingeleffer. 1993. Effect of minimal pruning and virus inoculation on the carbohydrate and nitrogen accumulation in Cabernet franc vines. Am. J. Enol. Vitic. 44(1):81-85. Savage, SD. and MA. Sall. 1983. Botrytis bunch rot of grapes: The influence of selected cultural practices on infection under California conditions. Plant Dis. 67: 77 1-774. Savage, SD. and MA. Sall. 1984. Botrytis bunch rot of grapes: Influence of trellis type and canopy microclimate. Phytopathology. 74:65-70. Scholefield, PB. and RC. Ward. 1975. Scanning electron microscopy of the developmental stages of the Sultana inflorescence. Vitis. 14: 14-19. . Scholefield, P.B., T.F. Neales, and P. May. 1978. Carbon balance of the Sultana vine (Vitis vinifera L.) and the effects of autumn defoliation by harvest-pruning. Aust. J. Plant Physiol. 5:561-570. Shaulis, N., and R. Smart. 1974. Grapevine canopies: management microclimate and yield responses. In: Proc. XIXth International Hort. Congress. pp. 255-265. 11-18 Sept, Warsaw. Sims, C.A., R.P. Johnson, and RP. Bates. 1990. Effects of mechanical pruning on the yield and quality of Muscadine grapes. Am. J. Enol. Vitic. 41(4):273-276. Skene, K.G.M. 1969. A comparison of the effects of “Cycocel” and tipping on fruit set in Vitis vinifera . Austr. J. Biol. Sci. 22: 1305-131 1. Smart, RE. 1973. Sunlight interception by vineyards. Am. J. Enol. Vitic. 24(4): 141-147. Smart, R.E. 1974a. Photosynthesis by grapevine canopies. J. Applied Ecology 1 1:997- 1006. Smart, R.E., NJ. Shaulis, and ER. Lemon. 1982b. The effect of Concord vineyard microclimate on yield. 2. The interrelations between microclimate and yield expression. Am. J. Enol. Vitic. 33:109-116. Smart, RE. 1984. Some aspects of climate canopy microclimate, vine physiology and wine quality. In: Proc. First Intl. Symp. Cool Climate Vitic. Enol. D.A. Heatherbell, P.B. Lombard, F,W. Bodyfelt, and SF. Price (eds.). pp 1-18. June 1984, Eugene, Oregon. Oregon State Univ. 85. 89. 91. 95. 20 . Smart, RE. 1985. Principles of grapevine canopy microclimate manipulation with implications for yield and quality. A review. Am. J. Enol. Vitic. 36(3):230-239. . Smart, RE. 1987. Influences of light on composition and quality of grapes. Acta Horticulture. 206237-47. Smart, R.E., J.K. Dick, I.M. Gravett and BM. Fisher. 1990. Canopy management to improve grape yield and wine quality - principle and practices. S. Afric. J. Enol. Vitic. 11:3-17. Srinivasan, C. and MG. Mullins. 1981. Physiology of flowering in the grapevine: A review. Am. J.Enol. Vitic. 32(1):47-63. Stergios, B.G. and GS. Howell. 1977. Effects of defoliation, trellis height, and cropping stress on the cold hardiness of Concord grapevines. Am. J. Enol. Vitic. 28( l):34»42. Tukey, L.D. 1957. Effects of controlled temperature following bloom on berry development of the Concord grape. Proc. Am. Soc. Sci. 71: 157-166. Weaver, R.J. 1958. Effect of leaf to shoot ratio on fruit quality and shoot development in Cari gnane and Zinfandel wine grapes. Am. Soc. Enol. Vitic. 14:1-12. Weaver, R.J., S.B. McCune, and CR. Hale. 1962. Effect of plant regulators on fruit set and beny development in certain seeded and seedless varieties of Vitis vinifera. Vitis 3:84-96. Weaver, R.J., W. Shindy, and W.M. Kliewer. 1969. Growth regulator induces movement of photosynthate products into fruits of ‘Black Corinth’ grapes. Plant. Physiol. 44:183-188. Weaver, R.J. and RM. Pool. 1968. Effect of various levels of cropping on Vitis vinifera grapevines. Am. J. Enol. Vitic. 19: 185-193. Weaver, R.J. and RM. Pool. 1971b. Chemical thinning of grape clusters (Vitis vinifera L.). Vitis. 10:201-209. Winkler, A.J. 1929. The effect of dormant pruning on the carbohydrate metabolism of the vine. Hilgardia. 4: 153-173. . Winkler, AJ. 1930. The relation of number of leaves to size and quality of table grapes. Proc. Am. Soc. Hrt. Sci. 27:158-160. Winkler, A.J. 1958. The relation of leaf area and climate to vine performance and grape quality. Amer. J. Enol. 9: 10-23. Winkler, A.J., J.A. Cook, W.M. Kliewer and LA. Lider. 1981. General Viticulture. University of California Press, Berkeley. 21 96. Wolpert, J.A., G.S. Howell, T.K. Mansfield. 1983. Sampling Vidal blanc grapes: 1. Effect of training system, pruning severity, shoot exposure, shoot origin, and cluster thinning on cluster weight and fruit quality. Amer. J. Enol. Vit. 34:72-76. 97. Wolpert,J.A. and GS. Howell. 1985. Cold acclimation of Concord grapevines. 1. Variation in cold hardiness within the canopy. Am. J. Enol. Vitic. 41:330-341. 98. Yang, Y.S. and Y. Hori. 1979. Studies on retranslocation of accumulated assimilates in ‘Delaware’ grapevines. I. Retranslocation of 14C-assimilates in the following spring after 14C feeding in summer and autumn. Tohoku J. of Agr. Res. 30(2):43-56. 99. Yang, Y.S., Y. Hori, and R. Ogata. 1980. Studies on retranslocation of accumulated assimilates in ‘Delaware’ grapevines. II. Retranslocation of assimilates accumulated during the previous growing season. Tohoku J. of Agr. Res. 31(2): 109-119. CHAPTER I INFLUENCE OF CANOPY CONFIGURATION ON CANOPY DEVELOPMENT, YIELD, FRUIT COMPOSITION, AND CANE QUALITY OF SEYVAL GRAPEVINES 22 Abstract Seyval grapevines were trained to three canopy configurations in 1988. One treatment filled the trellis (Full Trellis) early by spacing 45 buds along the cordon. Another simulated a hedge (Hedge) by hand pruning four inches around the cordon. A control was used where vines were balanced pruned to 8 nodes per 454 g. cane prunings. All vines were flower cluster thinned to 15 clusters per 454 g. cane prunings. Vine size and yield indices, as well as fruit quality, were evaluated each year. Canopy development was measured using point quadrat analysis taken during the 1991-1993 growing seasons. Cold hardiness evaluations were made in March of 1992-1994. The full trellis and hedge treatments had a larger number of nodes retained and earlier filling of the trellis area with foliage. However, hedging produced poorly spaced shoots that were less winter hardy. The full trellis treatment increased yield as a function of increased berries per cluster and cluster weight, as well as cane cold hardiness. Hedging increased yield by developing more clusters after thinning, but produced a higher incidence of Botrytis rot. There were no other differences in fruit quality between treatments. A canopy structure that fills the trellis early with properly spaced shoots appears to be the best canopy configuration for Seyval grapevines. Introduction Seyval is an important white wine grape cultivar across the Eastem United States. It is cold hardy, disease resistant, and phylloxera resistant. It does, however, form large, compact clusters susceptible to cluster rot under Michigan conditions and the large clusters can lead to overcropping. Crop levels are usually maintained by severe balance pruning and flower-cluster thinning. Reynolds et al. (12) suggested an optimal cropping level of 17 clusters per 500 grams of cane prunings in New York state. Severe pruning reduces the thinning expense incurred by the grower, but results in a later filling of the canopy and thus potential loss of photosynthesis. The current suggested balanced pruning formula is 15+10 (15 nodes retained for the first 454 gm of dormant one year cane prunings, and 10 nodes retained for each additional 454 gm). This results in a canopy that does not fill the trellis area until after fruit set Grapevine canopy development and density have been areas of recent interest and research. Spring shoot growth is dependant on the remobilization of carbohydrates stored in roots, trunks, cordons, and canes from the previous year (31,32), and cluster development is slow until leaf area is established (30). Shoots become net exporters of carbohydrates during anthesis (11). Cell division before anthesis and cell expansion after anthesis are the major determinants of the increase in weight of grape berries (3). Thus, it seems important to fill the trellis area with vegetative growth and leaf surface as early as possible to facilitate vine growth and fruit development. Increasing the number of nodes retained at pruning is the most common method used to establish an early full canopy (22). There are potential negatives. Increased shoot numbers can form a dense canopy that reduces light levels to 1% of the solar radiation in the fruiting zone (21,22). Light penetration into the canopy is important for photosynthesis 24 25 (20), fruitfulness (6,18), flower bud formation (18,21), fruit set (2,16), fruit quality (13,15,23), yields (18), and wood hardiness (5,18,26). It appears that an optimum leaf area exists above which light penetration into the canopy is restricted to the point that vine productivity begins to diminish, and below which the canopy is ineffective as a light trap for photosynthesis. A method to quickly establish and maintain this optimum leaf area to maximize the amount of assimilate produced in the vegetative season, and maximize vine productivity, fruit quality, and composition, plus insure vine growth and cold hardiness needs to be investigated for a large f ruited, overproducing cultivar such as Seyval. Mechanical hedging, selective hand pruning, or no pruning at all, increase the number of nodes retained for potential shoot production from the previous years vegetative growth, but, increasing shoot number increases canopy density (4). These minimally pruning management techniques also change vine morphology. They increase the leaf area before and decrease leaf production after bloom (4). Minimally pruned vines partition fixed carbon differently than conventionally pruned vines (4). They store more carbohydrates in old wood and less in their canes and roots (17). However, some of these methods have increased labor efficiency (19,25). Mechanical hedging has received increasing attention in the United States, since the pioneering efforts of the Australian researchers. This is suggested to be a cost effective alternative to hand pruning. Hedging leaves more nodes in the canopy area, increasing the amount of vegetation per foot of row and per square foot of canopy space (8,9,10). With no further crop control, increasing shoot number produces more clusters per vine, and thus greater sink competition. This in turn results in morphological characteristics such as reduced shoot length, leaf size, cluster size, cluster weight, and berry weight (1,7,29). This improves canopy microclimate (13,14) and increases photosynthetic efficiency through reduced shading (25). However, hedging alone has shown only to be useful for a few years before yields, fruitfulness, and fruit quality diminishes (8,9,10). Given the 26 differences in climate and limitations of sunlight and carbon assimilation between the North Eastern United States and Australia, an objective evaluation of the limitations of this technology deserves effort particularly for Seyval. Since thinning is required to control overcropping in Seyval, the crop level can be managed so that the influence of canopy configuration and decreasing pruning severity on canopy development, can be further investigated. Reducing labor expenses by mechanical hedging would greatly decrease the cost of producing this variety. [3.1:]... 1. To evaluate alternative pruning techniques that may allow expedient filling of the trellis for optimum fruit ripening in Seyval. 2. To determine the results of increasing bud numbers and balanced cropping (based on the previous years vine size) on canopy development, yield, fruit composition and quality in Seyval. 3. To evaluate balanced cropping with a set pruning number of buds retained, designed to space shoots at optimal distances from each other along the cordon in Seyval. , 4. To analyze the impacts of hedging in conjunction with balanced cropping by cluster thinning in Seyval. Materials and Methods PlamMaterial The experiment was established in March of 1988, 4.5 miles East of Lake Michigan (42-15’ latitude) at Fenn Valley vineyards in Fennville Mi. Mature, bearing Seyval grapevines planted in 1975 on an Oshtemo sandy-loam soil were used in the experiment. Rows were spaced 3.0 meters apart with vines 2.4 meters apart within rows. The vines were trained to a bilateral cordon on the top wire, 1.8 meters high (Hudson River Umbrella training system). Huninglreatments The three canopy configurations initiated in 1988 consisted of: 1) a full trellis treatment (FT) spacing 45 buds along the cordon with five-node canes and two-node spurs; 2) a mechanically pruned hedge (SH) was simulated by hand pruning four inches around the cordon; and 3) a control (C) (the current standard cultural method) was employed to ensure a typical rate of canopy development In all cases an 8+8 balance pruning formula (8 buds retained for every 454 grams of dormant one year cane prunings) was used. The control and full trellis vines were pruned to 45-nodes and the live one-year cane prunings weighed to estimate the previous year’s vine size. The node number for the control vines was then set for the duration of the growing season following the procedures stated above. Because hedged vines retained a larger number of nodes, a method was devised to estimate the weight of retained wood. After pruning the hedge treatment, the live canes were weighed and nodes counted. The remaining live nodes of canes on the vine were counted. A weight was assigned to the number of nodes over 45 retained on the vine, based on a node weight calculated in 1992 (113 gm per 52 nodes) and 1993 (113 gm per 50 nodes). This weight and the weight of cane prunings were added producing the vine 27 size estimate for each (SH) vine. Since pruning is not sufficient for optimal crop control in Seyval, cluster thinning was also required. The weight of live one-year-old canes at pruning (vine size) was used to determine cluster number after thinning. All vines were flower-cluster thinned just prior to or during bloom to 15-clusters per 454 grams of dormant cane prunings. At bud burst, nodes that did not produce a shoot (shootless nodes) were counted. Percent blind nodes, vegetativeness (vine size (gm) per shoot), count shoots (nodes retained at pruning minus blind nodes), and non-count shoots (total shoots minus count shoots) were calculated for each vine. Treatment comparisons of pruning data were analyzed each year and over years for the years applicable. II E l l E . : l' : Prior to harvest each year, Botrytis incidence assessments of each sample vine were taken. Vine rot and within cluster rot were subjectively evaluated using a one to five making where 1=0-20%; 2=21-40%; 3=41-60%; 4=6l-80%; and 5:81-100% of the clusters on the vine or benies within the cluster were rotten. Treatment vines were harvested each year within ten days of October 1. Each sample vine was hand harvested with the cluster number noted and the fruit placed into an individual bin. Each bin was then weighed in the field and a cluster weight calculated. Apical berries were taken from clusters of all sample vines within a treatment replication to create a 100 beny sample. From these samples pH, titratable acidity, soluble solids, and a berry weight were measured and the number of berries per cluster calculated. The number of non-count clusters per vine was calculated from the number of clusters harvested minus the number retained after bloom thinning. Treatment comparisons of harvest data were analyzed each year and over the years 1989,1991, 1992, and 1993. Using the harvest and pruning data, shoots per cluster (total shoots/ harvest cluster number), productivity (yield per previous 29 year’s vine size), and crop load (yield per current season’s vine size) were calculated for every sample vine and analyzed each year and over years. Fruitfulness (gm fruit per shoot) was calculated and analyzed during 1991-1993. Canopy Measrrrem enIS' Boint eradrat Analysis Point quadrat assessments were taken at three times during the growing season in 1991-1993. A thin metal rod was inserted into the canopy to determine the number of leaves above the fruiting zone (24). Five insertions along the cordon were made on each treatment vine at bloom, veraison, and harvest. Insertions were always made from the West side of the canopy. A frequency distribution of the number of contacts and a seasonal history of leaf layer formation were developed. ColdflardinessEvaluaticns Prior to pruning in 1993-1994, a cold hardiness evaluation was taken of the previous season’s vegetative growth. Previous work with Seyval (28) and a re-evaluation of that work (Appendix 1) has led to a definition of Seyval cane characteristics closely associated with cold hardiness. Classifications were created from hardiness analyses done on Dec 19, 1991 and Jan 22, 1992 on Seyval canes (Appendix 1). Periderm and primary bud T50’s were determined for variations within four cane characteristics, and over node positions along the cane. Canes having a medium diameter (7-10 mm), medium intemode length (6-8 cm), and dark brown periderm were found to have superior cane and primary bud cold hardiness. Persistent laterals and node position (2-9) did not have any influence on cold hardiness. Total one-year-old canes were counted on each treatment vine and each cane given an excellent, acceptable, or poor classification. Excellent canes consisted of ten or more live nodes with the optimal characteristics stated above. Acceptable canes had five or more living nodes but failed in one of the optimal characteristics. Canes with less than 30 five live nodes or failing at least two optimal characteristics were placed in the poor cold hardiness category. Samples of Seyval canes within the three classifications were collected in December, January, and March. Periderm and primary bud cold hardiness were significantly different among the treatments at the 0.01 level over the three dates. Excellent canes were superior while poor canes were inferior to acceptable canes. Using the three classifications was a rapid and effective way to evaluate the relative cold hardiness of each treatment vine. 13.”. I!I' A randomized complete block design was used with four blocks. Two rows, containing two blocks each, of self-rooted Seyval grapevines were utilized. Within each block, the treatments consisted of five sample vines. A guard row was situated outside and between the two treatment rows, and a guard vine separated each treatment Comparisons between treatments were done using the MST AT-C statistical computer package. Analysis was by ANOVA, with mean separations calculated using Duncan’s Multiple Range Test (27). Comparisons over years were made where applicable. Point quadrat frequency distributions were calculated using the MSTAT-C computer package. In the spring of 1990 a portion of the experiment was accidentally pruned by the grower, and thus, the treatments were analyzed using only two replications that year, and were not considered in the analysis over years. The experiment was terminated after pruning in March of 1994. Results and Discussion y' IT“ ”5.: .. E Vine yield was favored by increased node number retained. Both (FT) and (SH) vines had improved yields (25-30 percent; over a ton per acre) over the control (Table 1). 31 While (C) and (SH) yields fluctuated among years, the (FT) consistently produced 7 kg or more fruit per vine. Normally, these differences in yield might be the result of increased cluster number or cluster weight, composed of berry weight and number. Although large differences in nodes retained provided conditions favorable for large variations in cluster number, this variable was eliminated by the process of cluster thinning based on vine size. As a result, no significant differences were observed in cluster number at harvest (Table 2). The yield differences are, therefore, due to cluster weight and must be a result of variations in berry weight and/or the number of berries per cluster (Table 3). The (FT) system improved the components of yield compared to the other two treatments. Its average weight per cluster was the highest every year and 50 g heavier over years. Although occasional years (1989 and 1993) show a berry weight response, the major response is berry number per cluster. The (FT) treatment produced 10 percent more berries per cluster over years. Importantly, when both berry weight and beny number per cluster were significant, it was always for the (FT) treatment. It was surprising to note minimal differences in cluster weight of (C) and (SH) vines, contrasting previous work (1,7,29). Since the vines were flower cluster thinned near bloom each year, many more shoots per cluster existed in the (SH) canopy than the (C) at that stage. The greater leaf area at bloom may explain this response. The fnritfulness data suggest a reduced yield per node, but that was expected based on the thinned treatment response. Treatments did not affect soluble solids , pH, or titratable acidity (Table 4). The (C) berries appeared to accumulate the highest percent of soluble solids in each year, but were significantly greater only in 1989. The grams sugar per vine data are highly significant, but vary from year to year. Early, the best treatment is the (SH). However, over time it becomes clear that the better choice is the (FT) treatment. It produces the most sugar per vine over years. Interestingly, the (FT) also was the best at vegetative production and yield. Bottytis cinerea infection of the clusters was the most noticeable fruit quality variant among canopies (Table 5). Subjective ratings on the day of harvest, showed that the amount of Botrytis rot was severe across treatments every year. The average loss of crop was nearly 25 percent. Severe pruning produced adventitious shoots from secondary and tertiary buds, as well as vigorous laterals from basil nodes. This created dense microclimates immediately surrounding clusters of control canopies during fruit ripening. A dense canopy inhibits air flow and chemical penetration to the clusters. In 1989 and 1993 the higher amount of rot incidence in (C) clusters influenced yield. Rot within clusters was significantly lower for (FT) and (SH) in 1989, and for (FT) compared with (C) in 1991. The number of clusters with vine rot was significantly lower for (FT) and (SH) in 1989 and 1993. The (FT) treatment showed the least amount of vine rot in 1993. Over years, vine and cluster rot incidence were significantly lower for (PT) compared to (C). Clusters on the (FT) canopy consistently had less Botrytis rot infection. Proper shoot spacing created an optimal eanopy density limiting Botrytis rot infection in years of high incidence. if y . G l E Canopy configurations were achieved in the second year (1989). Although all the vines began the study at about 250-275 grams of cane prunings per meter of row, and yield varied by 30 percent, vine size differences were very small and seldom significant (Table 6). Vine size was similar among treatments in all of the years except in 1992. In that year, (FT) produced an average of 200 more grams of cane prunings per vine than the other two treatments. Averaged over years, the (FT) produced 60 grams more than (C) and 100 gm more than (SH). The (FT) treatment had the largest vine size and was also the treatment with the highest consistent yield. 33 There was a strong vegetative response by the vine to varying node number at the same cluster number. The full trellis and hedged systems increased node numbers within the canopy 4 and 10 fold, respectively from the control. Control vines produced the most vegetative growth as measured by both kilograms of cane prunings per node or shoot retained (Table 6). This supports Ruhl and Clingeleffer’s (l7) findings on carbohydrate partitioning of reserves, but was more likely due to two factors: 1) the stimulation of non- count shoot production by (C) vines, which resulted in two-thirds of the total shoot number per vine, from non-count positions; and 2) the competition for stored reserves, nutrients and water, among the large number of primary shoots arising from count positions in the other treatments, resulting in shootless nodes on (SH) vines. Shoot numbers among treatments did not vary as widely as the number of nodes retained due to adventitious shoot growth and shootless nodes. Adventitious shoots comprised much of the (C) canopy. These non-count shoots formed either along the cordon, or from secondary and tertiary buds when a primary shoot was present. Over years, (C) vines produced twice as many shoots from non-count sites as they did from count positions (Table 7). less than ten were produced each year by (FT) vines and very few produced by (SH). Whatever was gained by adventitious shoots on (SH) vines, was negated by the loss of count shoots during the growing season. Even taking into account the non-count shoot contribution, total shoot number was highly significant among treatments over years. The (C) vines produced the least amount of shoots and hedging the most. The large number of non-count shoots exemplifies the severe pruning practice of the (C) treatment. Another response to the differing bud populations per vine was the cane maturation and acclimation to cold. Cane quality varied with the number of nodes retained in the canopy (Table 8). Hedging produced the fewest number of excellent and the most poor quality canes each year. In 1993 (FT) vines matured significantly more excellent and 34 acceptable quality canes than the other two treatments. Over years, the treatments each matured significantly different numbers of excellent quality canes. The (FT) vines had the most (1 l) and (SH) vines the least (4) excellent quality canes. The (FT) vines had a greater number of acceptable canes and (SH) vines had a greater number of poor quality canes over the other two treatments. These data are significant whether expressed as direct data (Table 8) or as percentage of total canes per vine (data not shown). Whether data derived from canes can express accurately characteristics of older vine tissues is not known. However, this does pose a concern (not heretofore defined) about vine culture with systems of large bud numbers. With the canopies fully developed in the later years of the study, competition between shoots became evident. Nearly half of the (SH) buds failed to produce a shoot in 1992 and 1993 (Table 9), supporting the position that excess shoots can weaken a vine’s cold resistance. According to Smart (22), an optimal canopy has no more than three leaf layers. Point quadrat assessments taken during the 1991-1993 growing seasons, showed that increased shoot number per vine increased leaf layer number each year (Figs. 1, 2, and 3). The (SH) vines produced a greater average leaf layer number throughout the growing season each year. The (SH) canopies averaged 2 layers at bloom, and over 3 layers from veraison to harvest The (C) and (FT) canopies never obtained an average of 3 layers in any year. The (FT) canopies were also superior to (C) canopies each year. They continually had nearly half a leaf layer advantage. Most leaf layer production for (SI-I) canopies occurred between bloom and veraison, while (C) and (FT) canopies increased their rate of leaf layer production from veraison to harvest The (C) vines never completely filled the canopy with foliage during any of the growing seasons (Figs. 4-12). Gaps in the canopy still existed at harvest The (SH) and (FT) vines filled the trellis area by veraison each year (Figs. 412). However, (SH) canopies formed shaded areas at bloom each year. Hedging improved canopy formation 35 early in the growing season, consistent with prior work on hedging systems (8,9,10). However, increased shading occurred after veraison each year, contradictory to previous results with hedging systems (13,14,25). The upright growth of small shoots in hedged canopies caused the added leaf layer formation. The long, well developed shoots of control and full trellis canopies angled below the cordon later in the growing season exposing fruit and older leaves, and avoided leaf layer formation above the fruiting zone. The full trellis canopy was superior because it increased filling of the trellis with vegetation early in the growing season and allowed for further leaf development during ripening without substantial shading. Conclusions The treatments established three unique canopy architectures. Severe pruning and hedging were compared with a constant pruning level created to optimally space shoots along the cordon. Cluster number was consistent among treatments due to balanced cropping based on vine size. Thus, yield responses reflected differences in cluster weights. The full trellis and hedged systems increased yield by increasing cluster weight resulting from greater numbers of berries per cluster. Although the hedged canopies contained greater node numbers, the response of increased berry number was greatest in the full trellis treatment Retaining more nodes than currently recommended at pruning, increased shoot number within the canopy. With more shoots, canopies filled the trellis area with foliage earlier in the growing season. Creating a canopy that spaced shoots at optimum distances, increased light penetration and cane maturity, resulting in greater cane cold hardiness and deterrence to rot Hedging increased canopy formation before bloom, but shading became a problem after veraison. With poor light levels within the canopy, hedged vines ripened 36 fruit at the expense of vegetation and poor cane cold hardiness resulted. Dead wood accumulated in the canopy area and optimal spacing declined each year. Because of poor wood maturity, hedged vines would also be more susceptible to environmental stresses and disease. If hedging is the preferred method of pruning in Seyval, then cordon renewal is suggested for maintenance of vine health after four to five years. No advantages to vine growth were observed in the hedging system compared to the full trellis canopy. Thus, optimal shoot spacing should be the major consideration during pruning. Literature Cited 1. Bravdo, B., Y. Hepner, C. Loinger, S. Cohen, and H. Tabacman. 1984. Effect of crop level in a high yielding Cari gnane vineyard on growth, yield, and wine quality. Am. J. Enol. Vitic. 35:247-252. 1° Coombe, B.G. 1962. The effect of removing leaves, flowers and shoot tips on fruit set of Vitis vinifera L. J. Hort. Sci. 37:1-15. 3. Coombe, B.G. 1976. The development of fleshy fruits. Ann. Rev. Plant Physiol. 27(2):507-528. 4. Downton, W.J.S. and W.J.R. Grant. 1992. Photosynthetic physiology of spur pruned and minimal pruned grapevines. Aust. J. Plant Physiol. 19:309-316. 5. Howell, G.S., B.S. Stergios and SS. Stackhouse. 1978. Interrelation of productivity and cold hardiness of Concord grapevines. Am. J. Enol. Vitic. 29:187-191. 6. Howell, G.S., D.P. Miller, C.E. Edson and R.K. Streigler. 1991. Influence of training system and pruning severity on yield, vine size, and fruit maturity of Vignoles grapevines. Amer. J. Enol. Vitic. 42(3):191-198. 7. Kaps, ML. and G.A. Cahoon. 1989. Berry thinning and cluster thinning influence on vegetative growth, yield, fruit composition, and net photosynthesis of Seyval blanc grapevines. J. Amer. Soc. Hort. Sci. 114(1):20-24. 8. Morris, J.R., D.L. Cawthon, and J.W. Flemming. 1975. Effect of mechanical pruning on yield and quality of Concord grapevines. Ark. Farm Res. pub. 24(3): 1-12. 9. Morris, JR. and D.L. Cawthon. 1981. Yield and quality response of Concord grapes to mechanized pruning. Am. J. Enol. Vitic. 32(4):280-282. 10. Pollock, J.G., E.S. Shepardson, NJ. Shaulis, and DE. Crowe. 1977. Mechanical pruning of American hybrid grapevines. Trans. ASAE 20(5):817-821. 11. Quinlan, JD. and R.J. Weaver. 1970. Modification of pattern of the photosynthate movement within and between shoots of Vitis vinifera L. Plant Physiol. 46:527-530. 12. Reynolds, A.G., R.M. Pool and LR. Mattick. 1986a. Effect of shoot density and crop control on growth, yield, fruit composition and wine quality of ‘Seyval blanc’ grapevines. J. Amer. Soc. Hort Sci. 111(1):55-63. 13. Reynolds, A.G., and DA. Wardle. 1989a. Impact of various canopy management techniques on growth, yield, fruit composition and wine quality of Gewurtztraminer. Am. J. Enol. Vitic. 40: 121-129. 37 38 14. Reynolds, AG. and DA. Wardle. 1989b. Effects of timing and severity of summer hedging on growth, yield, fruit composition, and canopy characteristics of ‘De Chaunac’ 1. Canopy characteristics and growth parameters. Amer. J. Enol. Vitic. 40: 109-120. 15. Reynolds, AG. and DA. Wardle. 1989c. Effects of timing and severity of summer hedging on growth, yield,fruit composition, and canopy characteristics of ‘De Chaunac’ 2. Yield and fruit composition. Am. J. Enol. Vitic. 40:299-308. 16. Roublelakis, K.A. and W.M. Kliewer. 1976. Influence of light intensity and growth regulators on fruit set and ovule fertilization in grape cultivars under low temprature conditions. Am. J. Enol. Vitic. 27(4):163-167. l7. Ruhl,E.H. and PR. Clingeleffer. 1993. Effect of minimal pruning and virus inoculation on the carbohydrate and nitrogen accumulation in Cabernet franc vines. Am. J. Enol. Vitic. 44(1):81-85. 18. Shaulis, N., and R. Smart. 1974. Grapevine canopies: management microclimate and yield responses. In: Proc. XIXth International Hort Congress. pp. 255-265. 11-18 Sept, Warsaw. I9. Sims, C.A., R.P. Johnson, and RP. Bates. 1990. Effects of mechanical pruning on the yield and quality of Muscadine grapes. Am. J. Enol. Vitic. 41(4):273-276. 20. Smart, R.E. 1974a. Photosynthesis by grapevine canopies. J. Applied Ecology 11997-1006. 21. Smart, R.E. 1984. Some aspects of climate canopy microclimate, vine physiology and wine quality. In: Proc. First Intl. Symp. Cool Climate Vitic. Enol. D.A. Heatherbell, P.B. Lombard, F,W. Bodyfelt, and SF. Price (eds.). pp 1-18. June 1984, Eugene, Oregon. Oregon State Univ. 22. Smart, RE. 1985. Principles of grapevine canopy microclimate manipulation with implications for yield and quality. A review. Am. J. Enol. Vitic. 36(3):230-239. 23. Smart, R.E. 1987. Influences of light on composition and quality of grapes. Acta Hortic. 206:37-47. 24. Smart, R-E., C.M. Seaver. 1988. Handbook for Canopy Management Workshop. Pesented at 2nd International Cool climate Viticulture and Oenology Symposium, Matua Valley Vineyard Kumeu, New Zealand. (Jan. 1988). pg. 1-20. 25- Smart, R- E., J.K. Dick, I.M. Gravett and BM. Fisher. 1990. Canopy management to improve grape yield and wine quality - principle and practices. S. Afric. J. Enol. Vitic. 11:3-17. 26- Stergios. B.G. and cs. Howell. 1977. Effects of defoliation, trellis height, and cropping stress on the cold hardiness of Concord grapevines. Am. J. Enol. Vitic. 28(1):34-42. 30. 31. 39 Steel,R.G.D. and J.H. Torrie. 1980. Principles and Procedures of Statistics: A Biometric Approach. McGraw-Hill Book Company, New York, NY. Striegler, R.K., and GS. Howell. 1991. The influence of rootstock on the cold hardiness of Seyval grapevines: 1. Primary and secondary effects on growth, canopy development, yield, fruit quality and cold hardiness. Vitis. 30: 1-10. Weaver, RJ. and RM. Pool. 1968. Effect of various levels of cropping on Vitis vinifera grapevines. Am. J. Enol. Vitic. 19:185-193. Winkler, A.J., J.A. Cook, W.M. Kliewer and LA. Lider. 1981. General Viticulture. University of California Press, Berkeley. Yang, Y.S. and Y. Hori. 1979. Studies on retranslocation of accumulated assimilates in ‘Delaware’ grapevines. I. Retranslocation of l4C-assimilates in the following spring after 14C feeding in summer and autumn. Tohoku J. of Agr. Res. 30(2):43-56. . Yang, Y.S., Y. Hori, and R. Ogata. 1980. Studies on retranslocation of accumulated assimilates in ‘Delaware’ grapevines. II. Retranslocation of assimilates accumulated during the previous growing season. Tohoku J. of A gr. Res. 31(2):109-119. $8.093”. 03:35. 302 9:00:30 9:0: 0:523 553. 3:233 0:022 .85 sausage 6: 3 5e .4... C so a 232:2» 822, n. 00.00.30. 02 >_co cc 0000.. 0.0 300 000:» .009 .0 e805. c. .03 .0 use—ca 3:3 2 02". 0.038005 0.00 ~ 605.22 000520.020 m; 9:520. 05> 3:25 0+0 :0 to. 80.020 .0 .025: 05 8 9.6.0000 .3 003 .05. no.0 .5200 05 Eat «:60. :05 v 0 02050 00:00 02662 50.23 200:0 9:30: 0:0... 9.0: 02238800 003 20.500... 090: > .0053! 0095.082". m4 @5520. 05> 3:33 0+0 :0 .2 28020 .o .382. 05 2 9.2080 .00 003 .30. no.0 0:333: no 532 2 e055 9.03 «05> 2:0; :9... x 60520.. 0025.202“. m4 2 00:55 .2020 .030: 0:0 0+0 8 00:33 00:3 .9500 3 .80?va 0:0 80.. 9.0% 05 cc 0000.. 0.0 0:008 002C > .. . me me ~ .. me n em a m: a So so.» a: we a 2.0 0.: s 3222 .3 ones: a 82 o «to 8.0 no to e 3.3 m: x as «an; :3 e no.» o «3. a: 3.0 03 a so.» 2g. 3 8+8 .9280 > 32>.me $529; $529; 40529; €523 3539; 3539.4 335.2 amaze x 22> 22> 22> 22> 22> 2o; 2253: t 82 we? 53 one. one. one. 0210:; >0=0> 5.0“. “0 00:32.06 _0>>0m .0 003-003 :8: 20; 05> co 8:05.00: 9555 Lo 00:03:... .w mam]: 41 .005 00:0... 0.00.0.2 302 0:00:00 0:.00 030. 55.3 :0..0.0000 c005. .0... 28.5.5.0 .2. .o 5.. .0. .0. 80 o... .0 202......» 0020.. n. 00.02.00. 03. >50 :0 00000 0.0 0.00 000:5 .000. .0 00.0.2 5 .0.0 .0 05:30 .0500 0. 000 0.0.0585 0.00 N 0007.00. 0:0 000. 0.00.. 0:. :0 00000 0.0 0:005 000:... > 000590.020 0.. 0550.0. 05> 00:0.0 0+0 :0 .0. 0.0.00.0 .0 .0050: 0.... 0. 050.0000 .00 003 .0>0. 00.0 500.00 0:. 50.. 00.00. :05 v 0 00.0.00 00:00 00.650. :0.:3 0.00:0 0:0... 0500 00:0..050000 003 505.00.. 0000... x 000503.020 0.. 0550.0. 05> 00:0.0 0+0 :0 .0. 0.0.00.0 .0 .0050: 0:. 0. 050.0000 .00 003 .03. 00.0 0520000: 00 50.0. 0. 00:20 0.03 005> 050.... :0”. 3 0050.0. 000E0.0.00.0 0.. 0. 00:5... .2050 .030: 0:0 0+0 0. 00:0.0 0.03 00:.> .0..:00 > m: 00 00 .0 .. x .. a 00. a me o I n m: 00 00 cm .. 000. one o 2 000. 0.. mm C 0. .. a .3 n .9 o 0. 000. m... 00 mm 0. . . a 0. n 0.. o 0 .00. N 00 00 2. ~ .0. 3. mm 000. m: 8 om 0. . . a 00 a .3. o 0 000. 2. 00 mm .0 m: 0. 0.. 00 000. n. 0.0 0 .2020 0 082 0 .2020 0 082 0 .2020 0 082 x 008: a. 0.0.. .3“. > .228 .00.0>0:.> >0..0> ::0u. .0 00030020 .0100 .0 0007000. 50.. .0020: .0 .0050: .3050 0:0 0500.0 .0 .0050: 000: :0 2:05.00: 05:20 .0 00:05.5 .0 m..m50 :0 00000 0.0 0.00 0000.: .000. .0 00.05. 5 .0.0 .0 0500.0 .0500 0. 000 0.205005 0.00 N 0007.00. 0:0 000— 0.0g 00. :0 00000 0.0 0:005 000.: > 000505.020 0.. 0550.0. 05> 0000.0 0+0 :0 .0. 0.0.00.0 .0 .0050: 00. 0. 050.0000 .00 003 .0>0. 00.0 000.00 00. 50.. 00.00. 00:. v 0 00.0.00 00:00 00>050. 00.03 0.0000 0:00 0500 00000050000 003 505.00.. 0000... x 000520.050 0.. 0550.0. 05> 0000.0 0+0 :0 .0. 0.0.00.0 .0 .0050: 00. 0. 050.0000 .00 003 .0>0. 00.0 0520000: 00 50.0. 0. 0000.0 0.03 005.. 050... =0“. 3 0050.0. 000520.050 0.. 0. 00:50. .2050 .030: 0:0 0+0 0. 00:50 0.03 005> .9500 > .. e. 0.: a 00. a 03 0: 8.0 0.0 00.0 > x .. n 80 a 0.0 a New I 2 0.: v C 00 . n .0.. a 000 n 00.. .. a. NE a 000 n 8. 30. 0: 00. 00. 00. 2. 00.. m . .0 00.. . o .00 m .3. n 80 ~00. . a a: a mom a 8. m: 00.. 00.. 6.0 . a 80 a 03 a $0 .00. ~ 0 . . 00. . .0. N 8.0 8.0 00.. N 000 :0 $0 80. .. am 0.: a .0. a 00 . a. 0.0 a 30 0 000 . a 80 a 30 a 80 000. u. .00 .2020 5.0. .50. .0020 5.0. .50. .2020 5.0. 5.0. 00:50 .3 >000 .3 .2050 00.000 .3 >000 .3 .2050 000.00 .3 500 .3 0.00.0 x 0W00I 3 0.._0..P =0“. > .5500 .00.0>05> >0..0> :00”. .0 00:.>000.0 .0100 .0 0000.000. 50.. .0020: .0 .0.00.0.00...00 0:0 ...3 >000 ...3 3.00.0 :0 2:05.02. 0500.0 .0 00:00:... .0 .303. 43 .00. 00:00 25.3: :02 0.08000 058 0:2 50..; 8.0.0000 :00: .0.... 38.....0.» .8 .o .0.. 0... .C .00 0... .0 00250.0 08.0.. n. 3.02.00. 2.. 08 S 0000.. 0.0 0.0.. 80... .80. .0 020.2 s .20 .o 8.00... 0.00.. o. 0.... 0.0.058... 0.00 0 .8208. 0:0 000. 200.. 0... E 088 0.0 0000... $0... > 00520.05 50. 00.... u 05> .00 .0000 050.0 x 0000.20.00... 0.. 00.0.0.0. 05> 0000.0 0+0 :0 .0. 0.0.00.0 .0 .0050: 00. 0. 006.0000 .00 003 .0>0. 00.0 000.00 00. 50.. 00.00. 000. v 0 00.0.00 00000 0.5050. 00.03 0.0000 0000 00.00 0002350000 003 505.00.. 0000... 3 000500.020 0.. 00.0.0.0. 05> 0000.0 0+0 :0 .0. 0.0.00.0 .0 .0050: 00. 0. 00.0.0000 .00 003 .0>0. 00.0 00.200000 0.. 0.0.0. 0. 0000.0 0.03 0003 0.00... :0“. > 0050.0. 000030.020 0.. 0. 0000.0. 0.00.0 .0300 0:0 0+0 0. 0000.0 0.03 00:3 .0030 0 .. a 0.00. 0 0.00. o 0.00 0.. .0... 8.0 00.0 0.. 0.... 00.0 0.0 0 x 0.. 0... 0.... 00. - .. 0 .00. 0 0.00. o 0.00 0.. 00.0 . .0... $0 0.. .0... .0... .0.... 0.. 0... 0.0. 0.0. 000. 0.. 0..... 0.0.. 0.0.. 0.. .0... 00.0 00.0 0.. 00.0 8.0 8.0 0.. 00. o... 0.0. 000. .. 00 0.00. 0 0.00. n 0.0... 0.. .0... 00.0 00... 0.. 00.0 B... 00... 0.. .00 0.0. .00. .00. 0 .00 0.0.. 0.0.0 0 Z... 0.... 0.... 0 00.0 .5... I... 0 0.0. 0.0. 0.00 000. .. 0 0.8. 0 0.00 a 0.00 0.. 00.. 0.... 00.. 0.. 00.0 00... No.0 . 0 0.0. 0 0.0. 0 .0. 000. u. 0.0 x .0000 .50 <. :0 000.. x .0000 0.0 E 0.. 00.... 0 E0 0.0 E :0 00.... s 0000: > 0...... =0. 0 .0008 00.0.8... 0...; 0:0“. .0 02.380... .0100 .0 08.08. .02. .8200 .0 0000000... .0000 05> 000 00.000058 00.. 00 8:05.02. 0500.0 .0 8000...... .0 .305 .60... 095“. 03222 302 953.50 9.8: 2.5.. 55.3 c3233 522 .35 28:53 .2. 3 .C x; .C 4.... 2.. a £855.» 822, 4068232 oz: >_co :0 Emma 2a «in 39:. .39 .o 5222 c. .03 .o 9.755 3:8 2 26 80363:. San dam—wamw 0cm GOQF 30> 0...: CO 00062 26 QCGOE 80:... 6.030.— 050; 05> 05 :0 2053.0 uO .v‘oOOruOOflm Ucfl ”38-8"”. ”$8-93 u<.o¢.o~um u$o~-ou. 22.3 95.52 m; a 8.2. 33.2.. 202.833 «a; .2 2.5 £on 203 9033.0 05 55:5 «2:3 .0 «voowéoum new ”ioowécuw ”$8-83 ”xovémuu ”38.0"? 22; 9:52 m; a as»: 8.333 202.833 «a; .2 3.3.0 555, .ouo:\2o~m:_o m4 9:522 05> 3:35 n+0 ca .2 223.0 .o .352: 0..: 2 @5283 “on ma; .0>0. 35 .828 2: E2. mama. soc. v a 222:0 3:3 335...: so...) 835 new; 9:2. 35.3583 «a! 225:: wane: 635223.". m; 95:22 05> non—ea o+o ca .8 223.0 .o .383: 05 2 9.6.88 .3 ma; .0>0. no.0 .3583: mv 532 o. 3:35 22> 35> 9:9; =3”. .3589 ouoEESmau m... 2 3:55 3320 .030: new 95 2 ocean 20! 35> .2200 u N > . na 3 a am a to > x . an «.m a 5 a ad I .. g mm o a; a «a m: o.~ 2.. ed 83 m: we «.9 m6 2 to o.~ m.~ «m3 2 .3 .1 «a . 8 3 n 3 a on 33 ~ 2 0.0 3 ~ «a 90 5. 82 .. a mm a R a 3 .. a va n 3 a we 33 n. .3 x .2". 3 “om x .8 3 8m x 8m 3 8m 05> .3520 05> .0320 05> .3330 > a8: a 2...: .3“. . .228 6213c; >o=u> econ. 3 3539.5 313.3 39.39 E2. 825:. .a 283.0 93.5 new 55.3 8. 0258 .o 8:029... co 8:253: 9555 .0 8:02.... .n meE. 45 .30... omcmm 03:32 392 $58.50 9.3: 252 55.3 5:833 582 .85 23.253 .0: .o A... $F .C $m 05 «a .58:sz 83.9 m $23.32 95 >25 :o 823 2w 2% oon. .89 .0 new: 5 «ca 3 9555 .523 2 can 229:8... Sac N 6878? Ba 83 2mg 05 :o 083 2a 289: 085 > 6852236 9F @5522 was 8:95 n+0 ca 8. 223.0 .0 .3E:: 05 2 9.2085 .8 was .26. no.0 .5200 on. E9. mans so... v a 0223 8:8 3362 :02: 235 9.2. as? 3:22.583 was .2953: omuor x 63:223.". m.F @5522 was 2:56 o+o ca .2 223.0 .0 39:5: 2: 2 9.283 «ow «a; .0>0. no.0 @5233: mv £22 2 3.53 22: 85> 9:2» =3“. 3 .8522 26520520 mF 2 02.55 8520 830: new o+o 2 3.55 22: 35> .9200 > . . 9 9 an vF a 0F .. on em. usv > x . n vvd a vmd nu med I . n N cm 3 a 3 . . o v n FF w Fm w: Bio med omd mmaF . . n F m FF w 3 . . o m n 0F m cm . 2 Fee a mud n and «2: m: o F v F om .. n o 9 3 a on we end vvd mmd meF .. o m n FF a mm m: and mvd no.0 2:: N x. v F mm F N Fed v0.0 wFF aomF an F m o N F N me mad Fm.o Fwd oomF n. gm 3 .8 $529: :3 a. $529: .3 :3 85%... Saga .3 So: .8 mum 82w .3 ova: .8 cam 89% Ba ova: 5a 35 £39m .mo> £390 .mo> 2.5 £320 dc> £320 .mo> 2.5 £320 .25 £320 do> 0:5 x manor 3 «=3; =3“. > .9300 @8783 E9: 3330.3 on: 305 .8 5320 .wEgocs >c=u> 53m S .938 c. $3.39 E0: 9553 “a 3582 one: Ba 532m 93 on.» 05> :0 2:25am: 9.253 .0 00:022. .0 mqmfi. 03., macaw. 29:22 :62 @5850 9.3: 2.6. 55.3 5.3.33 522 .35 28:29.». .9. .o A... .x; .C «on 2.. a 28:29.». 329, u. 68.5.93... mF 3.53». 9..., 32:... n+0 cm .2 293.0 .o 59...... m5 2 9.6.88 .3 mm; .92 no.0 .898 9.. ES. menu. so... v a 3650 «2.8 3.68m. 50...; 23% 2.9. 9...»: 8223688 on; 20.53.. ouuo: n 6358236 mF 3.52m. 05> umcaa o+o ca .2 9836 .o .325: on. 2 9.6.88 .3 mm; 55. no.0 .ocEmouoc 9 £92 2 no.5... 22, 85> m___o..F =3”. > .3532 30:35.36 mF o. 3:55 .0520 .252. 93 0+» 2 3:2... 22; 35> 6.2.00 x I m 8 n 9‘ 0 cm 2 n o n o a mm x I m Fm n ov o FF I .. m mu n av n vm . . o o n m a mm . . m mu a CV o m mmmF 2 a mo nu mv 2 mm 2 o o n v m E I m on an an n vF mmmF mc mv om mm . . n o an F u 3 2. m vv an on 9 S meF M. .9m 22°F 2.30.52 ano 3.3 230.52 .560 :32. 2500-52 2.30 N 025: > maofi :9. x .9200 $239.5 >23) cam“. E _m>>mw 5 89-39 E9. 5282, .23 3582.. c3238 .85.. co 2:953: 955... .o 09.02.... .F m._m<._. 47 .000... 0000K 200.05. 302 0.000000 00.00 030. 0.5.3 02.0.0000 000.2 .05 08:00.0 .8 .o A... 0.. .C 0.0 o... .0 580.000 820., u. 00.000.00.000 00050.0: 0.00 00.000 00. .0 0.2 .000. .0 00:0. .0 00000 02. m 00.... 000. .0 020.0000 00000 3:000 .000 n 00.000.00.000 00050.0: 0.00 .00500 05 .0 000 00:0. .00 00000 00.. 0.0E .0 m .0 000.0000 00000 3:000 0.00.0003. > 00.000.00.000 00050.0: 0.00 .06000 5.3 00000 00.. 0.08 .0 OF .0 000.0000 00000 3:000 00.003 x 000220.020 0F 00.0.0.0. 003 0000.0 0+0 :0 .0. 0.0.00.0 .0 .00050 05 0. 00.0.0000 .00 003 .000. 00.0 000.00 00. 0.0.. 00.00. 000. w 0 00.050 00000 00000.0. 00.03 0.0000 0000 00.00 0000:05300 003 0000000.. 0000... 3 000880.020 0F 00.0.0.0. 003 0000.0 0+0 00 .0. 0.0.00.0 .0 .00500 00. 0. 00.0.0000 .00 003 .0>0. 00.0 00.0.0000: 00 0.0.0. 0. 0000.0 0.03 000.0 0...0.... :0". > 00530. 0000.20.00? 0F 0. 0000.0. .2020 .030: 000 0+0 0. 0000.0 0.03 0003 .00000 0 .. 0 F0 0 0F 0 0F .. 0 FF 0 «F 0 FF x .. 0 v 0 FF 0 0 I .. 0 F0 0 3 0 NF . 0 0F 0 cm 0 0F .. 0 v 0 0F 0 0 000F .. 0 F0 0 0F 0 0F 00 0F NF FF .. 0 v 0 FF 0 0 ~00F .. 0 F0 0 0F 0 x. 00 0 0F 0 . o 0 0 0 0 0 FO0F n. 0.0 ~ .000 H 0.00.0002 x 20:83 N .000 4 0.000002 x .00..00xm N .000 » 0.00.0002 0. 20:00.6 3 0000: > 059.250 0 0.0000 00.00.30... 00:00) 000“. .0 .0100 0. 000739 0.0.. :0. .00. .000 00000000 00 3:000 0000 00 0.000002. 00.020 .0 0000000. .0 m..m<._. .00... 0000.... 0.00.3.2 302 0.000030 00.00 0.00. 0.0:... 02.0.0000 000.2 .00.. 03:00.0 .0.. .o .0.. 0... .0. 0.0 o... .0 080......» 0029. n. 0000.20.020 m.F 00.0.0.0. 00.0 0000.0 0+0 00 .0. 0.0.00.0 .0 80.000 00. 0. 00.0.0000 .00 00... .0>0. 00.0 000.00 0... .00.. 00.00. 000. v 0 00.0.30 00000 00>00.0. 00.0... 0.0000 0000 00.03 0000._00.0000 003 .000..00.. 0000... u .0000.0.0.03.0 0F 00.0.0.0. 003 0000.0 0+0 00 .0. 0.0.03.0 .0 30.030 0... 0. 00.0.0000 .00 00.... .0>0. 00.0 00.200000 0.. 0.0.0. 0. 0003.0 0.0... 000... 050.... :3“. > 00520. 0000.0.0.03_0 0F 0. 0000.... 3.00.0 .0300 000 0+0 0. 0000.0 0.03 000... .0..000 x .. .. .. .. .. .. 0.. .. 0.00 a 0.. a 0.. a 0+ 0 00 a 00 a F. 00 a 00 N 0000: n a. 00 0 F. n... 0 n. 00 0. a. 3:30.30 0 .F 0 m a 00 a v a m a. 00 a 0 x .2060 0. 2.2.. .x. 05>... 0. 05>... 0. 05>... 93> .26 000.9... 00 0. 00 0. F0 0. .0000: 00:0 .0023 00 0000203 00.0..00... >0_.0> 000“. .0 000759 .00.. .050 030 .000 .03.. 00.30000. 00030090 .0....00 0. 00000 00000000 .0 0000.030 00 0.000000: 0500.0 .0 0000000. .0 39...: 49 amen: _ coma.m> _ E85 0 who fl ”.9 n .I wmé mm. m 1N m N n rm $9 I h AEV ..|I. wmd 6V II. N v 395:: .26. Km. 3 nommmaxo .mEm>oc_> >m=m> Emu E 35536 31mm .0 mm? -59 Ea: 395% «558.9% 30:8 353% cc 8:923:00 >858 B Baum .F .9“. :oEomE En. 909:8 x .r' ,-" n‘ I ....... . .P I!) G) r ,7 .r .r r .. .r .r " - *- .9 -' n ‘ I‘ I. ‘. ’2 __.> ... 4. I. .- I I £9 3 EV I Q E O V . o a} Pr . . Q, .. .. ... ... . r... ... . .. ...... .... . . . f. .....1 r. 5...... r.... ...... .f/z... . .. ...... x. . ... .... ... .. I . .. . . . . . .. .. .. . ... .. . r... ...... .. . I. I... .. . .... .... I .r .. . f .. x .... ... .... I... ... .. r... ... .. r. .... u. . I. ... . . . ' .. ._ .. .. ... ... . .. .... 7.. ... .. . 7. J.K., Iv... 3.... ....- ..A.... I ... . . .... ... .. .. .. . .x . . .... .... .1 .. r J.,] xxx! . .- .... ... .. ... ..... .. ... .1. . .l .... x. 4 .. I .... .. I”... ...... . ... .. .... .... .. . .2 . .- Ti .... A .. r. 1 .. r ... f , .. x .. ... - II F ... . r. J . r ... ...t . I .... . . . . ... ... . .. .... . r . ..... .. .. .. . ... ... . T. u A { r I. I . . . . .... .3 ... ... . .... ...! . 1- !, .... J . r! .4 I: r .I. .. I .. . .. .. .. . , r... .....I.A . . r! . cl r! .... In... . . . . .1. K ... .... . .. II I F 7.; .... ... .. ... 1. .I . .1; ...... al.. ..a. a}! ., flint-C ...j 71”» punt. J . r04. ’1'! .4 an T r .. .. ... . . . . . ... .. ... .13.... r/...... I... .1 .r. ... .. .... ...: . 7... ..... r... .f. j .. r.. .... .4 .. v .. ... . .. .... .I .. .... v. ... ... r. .... ..+ ,. r!... I... I... on... .. ... ...! .... I f... 1 .. .....z... . .. I .. .. s .. ‘ .. . . v..- l-cvff‘ .. Mn! .. ' .... . .... ,. .. H I ,. . .. :8 r ...: ...: . I .. ... I. -./ . ' ... .... . . ... .... . r. .... .. . ... T , ........ u 8 .2 . . . ' .I .l I l 1 T T I II- I I .I 1 ID V. .388 9: E can a 3:398. 209:8 QoN .mcozmsgcoo 30:3 025 o. 855 $5535 _m>>mw .2 $635 Eon .3 883 .m 8539: as: new $23: 209:8 6 59:5: 05 6 m8 759 ES. 80905 cogntfifi 3:262“. .N .9". (0/0) Aouenbajd coEmmc. En. $0250 ID If a _. ‘ . e _.- . - . .-" 3. 1’. .... . _ 4‘ I ‘. I 2' - 2' n . n o . l‘ 1’ . -' . v" I.’ j u" I ' O ,— ’. . ‘2' .. .‘ I. .‘ .- .. ’ . . .lv’ ‘1. ID ... O N (0/0) AouanbeJd LO N 51 O CO $9 I F: I Q E TIIIITTII'IIIUIIUIIItIIFri—IIIIIrIUIIIIU .80ch m5 5 gun a 25352 202:8 oboN .mcozmszcoo >358 025 2 8.55 85536 826m .2 .2820 Eon B :85? a 3532: c5: 2:... «26m: 202:8 .0 82:3: 2: 8 39.59 .2 3855 5:39:36 55:62“. .m .9... :o_:om:_ Lon 2o2:oo m N _. o w m v :3 l2 2 (%) AouanbaJd L0 N _!fmv I F: I 6: @— mm 0v .>no:mo 05 :_ now a 2:328: 222:8 QoN .mcozfizmzcoo >358 09.5 2 oocan mo:_>on§m _m>>om :2 8526 ES 3 «moan: E oosmmoEcS: o:m wo>moc 202:8 :o 5:83: 2: B 89-59 :2 835% :ozonEmfi 3:269“: .v .9“. CHAPTER II INFLUENCE OF CANOPY CONFIGURATION ON SHOOT GROWTH, LEAF DEVELOPMENT, AND LEAF AREA OF SEYVAL GRAPEVINES 53 Abstract Seyval grapevines were trained to three canopy configurations in 1988. One treatment filled the trellis early (Full Trellis) by spacing 45 buds along the cordon. Another simulated a hedge (Hedge) by hand pruning four inches around the cordon. A control was used where vines were balanced pruned to 8 nodes per 454 g. cane pnmings. All vines were flower cluster thinned to 15 clusters per 454 g. cane prunings. Shoot measurements were taken throughout the 1993 growing season, including a leaf area measurement at harvest. The full trellis and hedge treatments had a larger number of nodes retained and increased shoot and leaf development early in the growing season. The full trellis vines provided higher yields by producing heavier clusters with larger benies. Full trellis and hedged vines also produced a greater leaf area through verasion. But, by harvest, severely pruned vines obtained a leaf area equal to the other treatments. This resulted from new leaf production from laterals. The full trellis vines also produced many new leaves during fmit ripening, however, hedging showed little leaf production after veraison. Severe pruning caused an increase in Botrytis bunch rot in 1993, but, no other significant differences in fruit quality or vine size were observed. A canopy structure that exhibits early shoot and leaf development with properly spaced shoots and yet allows for new leaf production during fruit ripening, appears to be the best canopy configuration for Seyval grapevines. Introduction Seyval is an important white wine grape cultivar across the Eastern United States. It has numerous desirable characteristics. It is cold hardy, disease resistant, and phylloxera resistant. It does, however, f onn large, compact clusters susceptible to cluster rot under 54 55 Michigan conditions and the large clusters can lead to overcropping. Crop levels are usually maintained by severe balance pruning and flower-cluster thinning. Reynolds et al. (19) suggested an optimal cropping level of 17 clusters per 500 grams of eane prunings in New York state. Severe pruning reduces the thinning expense incurred by the grower, but results in a later filling of the eanopy and thus potential loss of photosynthesis. The current suggested balanced pruning formula is 15+10 (15 nodes retained for the first 454 gm of dormant one year cane prunings, and 10 nodes retained for each additional 454 gm). This results in a canopy that does not fill the trellis area until after fruit set. The photosynthate produced during the vegetative season must drive the processes of vegetative growth, fruiting, and flower bud development for next years crop. Reserves must also be stored over the winter for cold hardiness, and initiation of growth and development until a new canopy is producing photosynthate the following spring. Equally as signifieant is the mobilization and transport of assimilate between different organs of the vine in order to carry out these processes. Grapevine shoot development has been an area of recent interest and research. Spring shoot growth is dependant on the remobilization of carbohydrates stored in roots, trunks, cordons, and canes from the previous year (38,39), and cluster development is slow until leaf area is established (37). Shoots become net exporters of earbohydrates during anthesis (18). Cell division before anthesis and cell expansion after anthesis are the major determinants of the increase in weight of grape berries (4). Thus, it seems important to fill the trellis area with vegetative growth and leaf surface as early as possible to facilitate vine growth and fruit development Leaf development is important for fruit and shoot maturity. A leaf number greater than 10 per shoot is necessary for optimal fruit ripening and shoot weight (34,36). Leaf photosynthesis is greatest when the leaf is fully expanded, and then gradually declines Shortly after ( 17). The amount of photosynthate produced by web leaf is important for the 56 development of the subtending bud (6,27). Over eighty percent of the assimilate a cluster collects is produced by the leaves on the same side of the shoot (14,15). Buttrose (2) found that as leaf number was reduced, root dry weight decreased most severely, followed by reductions in berry development, and then trunk and shoot carbohydrate reserves. Increasing the number of leaves per shoot increases berry sugar, weight, and coloration at harvest (34). However, the production of each new leaf adds to the vine leaf area and the formation of leaf layers within the canopy. The leaf area of the vine is critical because it produces the assimilate necessary for crop ripening and wood maturity. A minimum leaf area of 7-14 cm"2 per gram of fruit is needed for adequate fniit ripening (10,31). Smart showed that a dense canopy can reduce light levels to 1% of the solar radiation in the fruiting and renewal zone (28), and an optimal canopy has no more than three leaf layers (29). Light penetration into the canopy is important for photosynthesis (11,26), f ruitfulness (8,25), flower bud formation (25,28), fruit set (3,23), fruit quality (20,22,30), yields (25), and wood hardiness (7,25,32). Thus, it is evident that a maximum leaf area exists where light penetration into the canopy is restricted to the point that vine productivity begins to diminish. The maintenance of this leaf area is important to maximize the amount of assimilate produced in the vegetative season for vine growth and development. When this maximum leaf area to facilitate vine growth is established is of great interest. Mechanieal hedging, selective hand pruning, or no pruning at all, increase the number of buds retained for potential shoot production from the previous years vegetative growth. These minimally pruning management techniques change vine morphology. They increase the leaf area before bloom and decrease leaf production after (5). Minimally pruned vines partition fixed carbon differently than conventional pruned vines (5). They Store more carbohydrates in old wood and less in their canes and roots (24). Increasing Shoot number increases canopy density (7) and the labor expense incurred by cluster thinning fruitful varieties. Mechanical hedging has received increasing attention in the United States, since the pioneering efforts of the Australian researchers. This is suggested to be a cost effective alternative to hand pruning. Hedging leaves more buds in the canopy area, increasing the amount of vegetation per foot of row and per square foot of canopy space (12,13,16). With no further crop control, increasing shoot number yields more clusters per vine, and thus increased sink competition and higher crop levels. This in turn results in morphological characteristics such as reduced shoot length, leaf size, cluster size, cluster weight, and berry weight (1,9,35). Whole canopy microclimate improves (20,21) and photosynthetic efficiency increases through reduced shading (31). However, hedging alone has shown only to be useful for a few years before yields, fruitfulness, and fruit quality diminish (12,13,16). Given the differences in climate and limitations of sunlight and carbon assimilation between the North Eastern United States and Australia, an objective evaluation of the limitations of this technology deserves effort and Seyval would seem to be a good eandidate for this effort. Since thinning is required to control overcropping in Seyval, the crop level can be managed so that the influence of canopy configuration and decreasing pruning severity on shoot and leaf development, can be further investigated. Reducing labor expenses by mechanical hedging would greatly decrease the cost of producing this variety. E‘ l:l"' 1. To determine the results of increasing bud numbers and balanced cropping (based on the previous years vine size) on vegetative growth, yield, and fruit composition and quality in Seyval. 58 to To evaluate balanced cropping with a set pruning number of buds retained, designed to space shoots at optimal distances from each other along the cordon in Seyval. 3. To analyze the impacts of hedging in conjunction with balanced cropping by cluster thinning in Seyval. 4. To measure the impact of these pruning/cropping strategies on shoot, lateral, and leaf development in Seyval. 5. To measure the impact of these pruning/cropping strategies on leaf area formation in Seyval. Materials and Methods ElantMaterial The experiment was performed during the 1993 growing season, 4.5 miles East of Lake Michigan (42-15’ latitude) at Fenn Valley vineyards in Fennville Mi. Mature, bearing Seyval grapevines planted in 1975 on an Oshtemo sandy-loam soil were used in the experiment. Three unique canopy configurations were initiated in 1988. Rows were spaced 3.0 meters apart with vines 2.4 meters apart within rows. The vines were trained to a bilateral cordon on the top wire, 1.8 meters high (Hudson River Umbrella training system). Burningffreatments The three canopy configurations consisted of: 1) a full trellis treatment (FT) spacing 45 buds along the cordon with five-node canes and two-node spurs; 2) a mechanically pruned hedge (SH) was simulated by hand pruning four inches around the Gordon; and 3) a control (C) (the cunent standard cultural method) was employed to ensure a typical rate of canopy development In all cases an 8+8 balance pruning formula (8 buds retained for every 454 grams of dormant one year cane prunings) was used. 59 The control and full trellis vines were pruned to 45-nodes and the live one-year cane prunings weighed to estimate the previous year’s vine size. The node number for the control vines was then set for the duration of the growing season following the procedures stated above. Because hedged vines retained a larger number of nodes, a method was devised to estimate the weight of retained wood After pruning the hedge treatment, the live mnes were weighed and nodes counted. The remaining live nodes of canes on the vine were counted. A weight was assigned to the number of nodes over 45 retained on the vine, based on a node weight of 113 gm per 50 nodes. This weight and the weight of cane prunings were added producing the vine size estimate for each (SI-1) vine. Since pruning is not sufficient for optimal crop control in Seyval, cluster thinning was also required. The weight of live one—year-old canes at pruning (vine size) was used to determine cluster number after thinning. The treatments were flower-cluster thinned just prior to or during bloom to lS-clusters per 454 grams of dormant cane prunings. At bud burst, nodes that did not produce a shoot (shootless nodes) were counted. Percent blind nodes, vegetativeness (vine size (gm) per shoot), count shoots (nodes retained at pruning minus blind nodes), and non-count shoots (total shoots minus count shoots) were calculated for each vine. II E l l E . C l' ! Treatment vines were harvested October 1. Each sample vine was hand harvested, the cluster number recorded and the fruit placed into an individual bin Each bin was then weighed in the field and a cluster weight calculated. Apical berries were taken from clusters of all sample vines within a treatment replication to create a 100 berry sample. From these samples pH, titratable acidity, soluble solids, and a berry weight were measured and the number of berries per cluster calculated. Using the harvest and pruning data, shoots per cluster (total shoots/ lmvest cluster number) was calculated. ShootandieaLMeasurements During the 1993 growing season, detailed shoot and leaf measurements were taken on one vine from each treatment block The vines were selected to conform to an average vine size range (0.57-0.91 kg). Within each sample vine, six shoots were randomly flagged along the cordon for vegetative measurements during the growing season. These included: 1) primary shoot length (PSL); 2) primary shoot leaf number (PSLF); 3) lateral number; 4) lateral length (LL); and 5) lateral leaf number (LLF); measured on May 26 (WK3 weeks after bud burst), June 2 (WK4), June 9 (WKS), June 16 (WK6), June 30 (WK8), July 29 (WK12), and September 23 (WK21) one week prior to harvest. Additional calculated data included total leaves per shoot (T SLF=PSLF+LLF) and total vegetative shoot length (PSL+LL). Treatment comparisons were made at appropriate dates and over dates on those variables measured or calculated. At thinning, three of the sample shoots were defruited while one cluster was retained on the other three. Shoot development with and without a fruiting sink, was compared for the remainder of the growing season. A standard curve was developed to allow rapid determination of flower number per cluster based on cluster length. Thirty clusters were randomly sampled from guard vines just prior to bloom. Rachis length was calculated using the length of rachis from the lowest basal arm to the tip added to the length of the lowest basal arm. Flowers of each sample cluster were counted and a strong correlation was found (Fig. 1). The rachis length of clusters on sample shoots was measured and an estimated flower number calculated. At harvest, berry pedicles of sample shoot clusters were counted and the percentage of berries set analyzed. A count of shoots longer than 10 cm and bearing three or more leaves was taken two weeks prior to harvest. Primary leaves per vine (PSLF*SC), lateral leaves per vine (ILF‘SC), and total leaves per vine (T SLF‘SC) were then calculated and compared between treatments. A leaf area standard curve was developed for primary shoot and lateral 61 leaves at WK21 by sampling 100 primary shoot leaves and 50 lateral leaves randomly from guard vines. The mid-rib length and area was measured for each leaf. A strong curvilinear relationship was found between mid-rib length and leaf area for primary shoot (Fig. 2) and lateral leaves (Fig. 3). Mid-rib lengths of leaves from sample shoots were measured WK21 and leaf area calculated using the standards. Leaf area was calculated on a per leaf, per shoot, per vine, per cluster, and per yield basis for comparison. Leaf area accumulation per vine over the 1993 growing season was calculated using the average leaf area and number of both primary and lateral leaves in each treatment. E' 112' II' A randomized complete block design was used with four blocks. Two rows, containing two blocks each, of own rooted Seyval grapevines were utilized. Within each block, the treatments consisted of five sample vines. A guard row was situated outside and between the two treatment rows, and a guard vine separated each treatment. Comparisons between treatments were done using the MST AT-C statistical computer package. Analysis was by ANOVA, with mean separations calculated using Duncan’s Multiple Range Test (33). Leaf area regressions were calculated using the Delta Graph computer package. Results and Discussion If 1:. ll lE .: l' Vine yield was favored by a decrease in pruning severity (Table 1). The (FT) and (SH) vines produced 3540 percent more fruit than (C). Normally, these differences in 62 yield might be the result of increased cluster number or cluster weight. Although large differences in nodes retained provided conditions favorable for large variations in cluster number, this variable was eliminated by the process of cluster thinning based on vine size. As a result, no significant differences were observed in cluster number at harvest (Table 1). The yield differences are, therefore, due to cluster weight, and must be a result of variations in berry weight and/or the number of berries per cluster (Table 1). The (FT) system improved the components of yield compared to the other two treatments. Its average weight per cluster was 17-27 percent heavier. Since there were no significant differences in the percent berry set or number of benies per cluster among treatments, the major response was berry weight in 1993. The (FT) treatment produced berries nearly 10 percent greater than the other treatments. The minimal differences in cluster weight of (C) and (SH) vines contrast with previous work (1,9,35). Since the vines were flower cluster thinned near bloom each year, many more shoots per cluster existed in the (SH) canopy than the (C) at that stage. The greater leaf area at bloom may explain this response. The fruitfulness data suggest a reduced yield per node, but that was expected based on the thinned treatment response. Fruit composition data provided little basis for discrirrrinating among treatments (Table 1). There were no differences among soluble solids , pH, and titratable acidity. Botrytis cinerea infection of the clusters was the most noticeable fruit quality variant among canopies. Subjective ratings performed the day of harvest showed that the amount of Botrytis rot was severe across all treatments. The average loss of crop was nearly 25 percent. No significant differences were observed in rot incidence within clusters, however, the higher amount of rot incidence in (C) clusters expressing rot (vine rot), influenced yield (Table 1). The (FT) vines showed less incidence of vine rot in 1993. The full trellis canopy was the least conducive to Botrytis rot infection. 63 No significant differences were found in vine size; however, increasing the number of nodes retained reduced vegetative vine growth measured by kilograms of cane prunings per node (Table 2). This result was likely due to two factors: 1) the stimulation of non- count shoot production in (C) vines, which resulted in three—quarters of the total shoot number per vine from non-count positions; and 2) the competition for stored reserves, nutrients and water among the large number of primary shoots arising from count positions in the (SH) canopies resulted in shootless nodes. Adventitious shoots comprised much of the (C) canopy. These non—count shoots formed along the cordon, or from secondary and tertiary buds when a primary shoot was present. Since nearly all of the count shoots were needed to cany the crop load, non-count shoots served as additional sources of photosynthate. Some of these adventitious shoots produced flower clusters. Because they developed as quickly as on non-count shoots, they were thinned during bloom as well. The competition among shoots was most evident in the hedged vines, as many buds failed to produce any growth. Less than ten non-count shoots were produced by the (FT) vines and very few produced by (SH). Whatever was gained by adventitious shoots on (SH) vines, was negated by the loss of count shoots during the growing season. The (SH) vines also produced two to four times as many shootless nodes as the other treatments. Even taking into account the non-count shoot contribution and shootless nodes, total shoot number was highly significant among treatments (Table 2). The (C) vines produced the least amount of shoots and (SH) the most The large number of non-count shoots exemplifies the severe pruning practice of the (C) treatment; while the shootless nodes in hedged vines indicate vine capacity is attained. ShootAssessmems Because shoots begin exporting carbohydrates during anthesis, shoot development fin)“ him; no as!) $2: zfi_ \le Sfiggt hm Glut 64 early in the spring is important for overall vine production. Seyval clusters are very large and become strong carbohydrate sinks at fruit set. Berry growth and most of the vegetative growth in a season occurs between bloom and verasion (4). Laterals form just prior to bloom and grow for the remainder of the season. Shoots thinned of fruit are sources of photosynthate for various sinks within the vine. During the 1993 growing season, detailed shoot measurements were taken on six sample shoots from one vine in each treatment block. Shoots of (FT) vines were more developed before bloom. Between WK3 and WK6, (FT) vines had a greater primary shoot length than (C) or (SH) (Fig. 4). This suggests the (FT) shoots were able to utilize reserves better than the other treatments and become self sufficient earlier. The earlier shoot and leaf development of the (FT) vines coincides with early cell division in the beny (4), and may be the cause for their superior berry weights. The (C) shoots caught up in length to (FT) shoots by WK12, and they both averaged more than 40 cm longer than (SH) shoots at harvest. Shoot growth rate for each treatment increased after WK6 and decreased again after WK12. Little primary shoot growth was achieved by (SH) after veraison, while (C) and (FT) shoots continued to grow. Since intemode length remained consistent between treatments at every date, leaf production was closely related to primary shoot length. The (FT) shoots averaged one more leaf on WK4 (Fig. 5). At veraison, (C) and (FT) primary shoots averaged 5 more leaves than (SH). This response continued through harvest, but as (FT) and (SI-l) primary shoots stopped producing leaves, (C) primary shoots increased leaf number. New leaf growth rate for each treatment increased after WK6 and decreased again after WK12. Lateral growth began to appear during WK6. By WK12, (C) and (FT) shoots produced over twice as many laterals than (SH) (Fig. 6). None of the canopies produced new lateral shoots after verasion. The laterals of (C) and (FT) shoots produced more leaves than (SH) shoots by verasion, and continued to produce more till harvest (Fig. 7). 65 They also had a greater rate of elongation between WK6 and WK8, and length at harvest (Fig. 8). Rate of elongation after anthesis was inversely related to bud number. Shoots of (SH) vines were considerably smaller with less leaves and lateral growth. No differences were found in the total vegetative shoot length until WK21 when (SH) shoots were inferior to the other two treatments (Fig. 9). Shoots of (SH) also had fewer leaves per shoot than the (C) shoots on WK12 and WK21 (Fig. 10). Over dates measured, (SH) shoots were inferior to (C) and (FT) shoots in length and leaf number, while control and (FT) shoots were never significantly different from each other. This agrees with previous work with hedging systems (1,9,35). Shoots with and without a cluster were compared in each treatment and no significant differences or interactions were found at any date. However, shoot and leaf measurements on defruited shoots were significantly greater than shoots bearing fruit over the year. Canopy type influenced the rate of growth between veraison and harvest (Table 4). At veraison the (SH) vines averaged 2800 leaves, 600-800 more than (C) or (FT). Only 12 percent (389 leaves) of the total (SH) leaf production was grown between veraison and harvest. The (C) and (FT) vines produced 900 (31%) and 700 (24%) of their leaves during this time respectively. Lateral shoots were responsible for the additional leaf production after veraison. Laterals provided young, productive leaves during fruit ripening. The (C) and (FT) canopies produced laterals that continued to produce leaves after veraison, while (SH) shoots had little lateral leaf production. The lack of new leaf production after veraison in (SH) vines suggests older, less photosynthetically efficient (17) leaves were responsible for photosynthate production during fruit and wood ripening. Although, fruit quality was not affected, wood maturity was a concern. The vegetativeness per shoot was significantly lower in (SH) vines (Table 2). LeaLAreaMeasurements The mid-ribs of leaves from sample vines were measured at harvest. The (SH) primary shoot leaves were significantly smaller than those of the other two treatments (Table 3). However, no differences among treatments were seen in lateral leaf area or the whole shoot average area per leaf. Shoot leaf area was directly related to leaves per shoot and inversely related to shoot number (Table 3.). The (SH) vines produced the fewest leaves per shoot and the least leaf area per shoot. They produced a primary shoot leaf area over 750 square centimeters smaller than the other two treatments and a 70 percent decrease in lateral shoot leaf area from the control; leaf area per whole shoot was over 45 percent smaller than (C) and (FT). The (FT) and (C) vines never varied significantly in leaf number or leaf area per shoot. Obviously, there exists a limit in the number of vegetative sinks a vine can support and grow to full potential. Shoots with and without a cluster were compared in each treatment and no significant differences or interactions were found. Thus, the crop level influence was over the entire vine and not limited to the shoot. Leaf area per vine at harvest was calculated using the average area per leaf, total leaves per shoot, and number of shoots per sample vine (Table 4). Although (SH) shoots were less developed and possessed less leaf area, no significant differences among treatments in the total number of leaves or leaf area per vine at harvest were observed (Table 4). (SH) shoots averaged over 3000 leaves per vine with an area more than 16 meters square. However, it is interesting to note how the treatments achieved this leaf area. Using the average leaf area for primary shoot and lateral leaves and the leaf number of each of these at the dates counted, a seasonal history of leaf area development was constructed (Fig. 11). Increasing the number of shoots within the canopy increased the leaf area from pre-bloom until after veraison, supporting Downton and Grant’s results (5) on minimal pruned vines. Lateral shoots produced the additional leaf area in the (C) and (FT) vines from veraison to harvest. Leaf area per cluster was not significantly different among 67 treatments, although (FT) and (SH) had 1000 and 1500 square centimeters less per cluster than (C) respectively. Significant differences among treatments did surface when leaf area was expressed per gram of yield. The (FT) and (SH) vines produced over 10 cm"2 less leaf area per gm yield than (C) vines (Table 4). However, all the treatments produced over 20 cm"2 of leaf area per gram of fruit, sufficient to adequately ripen fruit (10,31). A greater leaf area, necessary to ripen fruit and wood in Michigan’s climate, was expected. Canopy configuration and increased bud numbers increased the amount of leaf area through veraison, though the leaf area response to increasing node number was limited. The (SH) vines never achieved greater amounts of leaf area compared to the (FT) vines. Conclusions The treatments established three unique canopy architectures. Severe pruning and hedging were compared with a constant pruning level created to loptimally space shoots along the cordon. Increasing node number at pruning increased yield with no affect on pH, titratable acidity, or soluble solids. Cluster number was consistent between treatments because of balanced cropping based on vine size. Yield responses were due to an increase in berry weight. The greatest cause for a decrease in fruit quality or crop loss was Botrytis rot infection. An average of 25 percent of the crop was lost between all of the treatments. The adventitious shoots forming from secondary and tertiary buds, as well as vigorous laterals from basil nodes, formed a dense leaf area immediately around the control clusters. A dense canopy inhibits air flow and chemical penetration to the clusters. Crop loss was most severe in control vines, while the full trellis canopy was the most consistent in avoiding rot infection due to optimally spaced shoots creating a low leaf density around the clusters. F- v, lit 68 It is evident that early shoot development is important for fruit development. Severe pruning inhibited shoot and leaf area development and promoted adventitious shoot growth early in the growing season. Because of greater shoot numbers, the hedged and full trellis canopies developed earlier. However, the full trellis canopy configuration was superior because of optimal vegetative sink competition and leaf area development allowing for better shoot development. This resulted in the full trellis vines yielding the heaviest clusters with the largest berries. Full trellis vines seldom producing shoots from non-count positions. In the hedging system, vines achieved their full capacity, since all of their vegetative growth arose from count positions and shoot abortion occurred throughout the growing season. Dead wood accumulated in the canopy and optimal shoot spacing declined each year. Because sufficient source levels were available, the control and full trellis canopies produced large primary shoot leaves and lateral shoots. The lateral leaves were an important component of total leaf area. However, much of the leaf and leaf area production by control canopies was produced after verasion. Small primary shoot leaves comprised most of the leaf area of hedged canopies. Full trellis and hedged vines produced a greater leaf area through verasion; however, new leaf production in hedged vines decreased because of the lack of laterals. Laterals provided young, productive leaves during fruit ripening. The full trellis canopy was superior because it increased shoot development early in the growing season and allowed for further leaf area development during fruit and wood ripening. This was not an advantage in 1993, since each treatment produced more than enough leaf area to sufficiently ripen fruit. However, if crop levels were to be increased, mid to late season leaf production could be crucial. Increasing node number per vine by hedging showed no advantage over the full trellis system in production of leaf area or fruit. Hand pruning to a set number of buds spaced along the cordon produced the best canopy configuration in this experiment. to 10. 11. l2. 13. Literature Cited . Bravdo, B., Y. Hepner, C. Loinger, S. Cohen, and H. Tabacman. 1984. Effect of crop level in a high yielding Cari gnane vineyard on growth, yield, and wine quality. Am. J. Enol. Vitic. 35:M7-252. ' Buttrose, MS 1968. The effect of varying berry number on Gordo grapevines with constant leaf area. Vitis 72299-302. . Coombe, B.G. 1962. The effect of removing leaves, flowers and shoot tips on fruit set of Vitis vinifera L. J. Hort. Sci. 37:1-15. Coombe, B.G. 1976. The development of fleshy fruits. Ann. Rev. Plant Physiol. 27(2):507—528. Downton, W.J.S. and W.J.R. Grant. 1992. Photosynthetic physiology of spur pruned and minimal pruned grapevines. Aust. J. Plant Physiol. 19:309-316. Hale, CR, and P.J. Weaver. 1962. The effect of developmental stage on direction of translocation of photosynthate in Vitis vinifera . Hilgardia 33289-131. Howell, G.S., B.S. Stergios and SS. Stackhouse. 1978. Interrelation of productivity and cold hardiness of Concord grapevines. Am. J. Enol. Vitic. 29:187-191. Howell, G.S., D.P. Miller, C.E. Edson and R.K. Streigler. 1991. Influence of training system and pruning severity on yield, vine size, and fruit maturity of Vignoles grapevines. Amer. J. Enol. Vitic. 42(3): 191-198. Kaps, M.L. and G.A. Cahoon. 1989. Berry thinning and cluster thinning influence on vegetative growth, yield, fruit composition, and net photosynthesis of Seyval blanc grapevines. J. Amer. Soc. Hort. Sci. 114(1):20-24. Kliewer, W.M. and R.J. Weaver. 1971. Effect of crop level and leaf area on growth, composition and correlation of ‘Tokay’ grapes. Am. J. Enol. Vitic. 23:172-177. Kriedemann, P.E., B.R. Loveys, and W.J.S. Downton. 1975. Internal control of stomatal physiology and photosynthesis. 2. Photosynthetic responses to phaseic acid. Aust. J. Plant Physiol. 22553-567. Morris, J.R., D.L. Cawthon, and J.W. Flemming. 1975. Effect of mechanical pruning on yield and quality of Concord grapevines. Ark. Farm Res. pub. 24(3): 1- Morris, J.R. and D.L. Cawthon. 1981. Yield and quality response of Concord grapes to mechanized pruning. Am. J. Enol. Vitic. 32(4):280-282. 69 14. .\1 15. .\1 19. R 31.11 14. 15. 16. 17. 18. 19. 20. 21. 25. 26. 7O Motomura, Y. 1990. Distrbution of 14C-assimilates from individual leaves on clusters in grape shoots. Am. J. Enol. Vitic. 41:306-312. Motomura, Y. 1993. 14C-assimilate partitioning in grapevine shoots: Effects of shoot pinching, girdling of shoot, and leaf -halving on assimilate partitioning from leaves into clusters. Am. J. Enol. Vitic. 4421-7. Pollock, J.G., E.S. Shepardson, N.J. Shaulis, and DE. Crowe. 1977. Mechanical pruning of American hybrid grapevines. Trans. ASAE 20(5):817-821. Poni, S., C. Intrieri, and O. Silvestroni. 1994. Interactions of leaf age, fruiting and exogenous cytokinins in Sangiovese grapevines under non-irri gated conditions. 1. Gas exchange. Am. J. Enol. Vitic. 45(1):71-78. Quinlan, J.D. and R.J. Weaver. 1970. Modification of pattern of the photosynthate movement within and between shoots of Vitis vinifera L. Plant Physiol. 46: 527-530. Reynolds, A.G., R.M. Pool and LR. Mattick. 1986a Effect of shoot density and crop control on growth, yield, fruit composition and wine quality of ‘Seyval blanc’ grapevines. J. Amer. Soc. Hort. Sci. 111(1):55-63. Reynolds, A.G., and DA. Wardle. 1989a. Impact of various canopy management techniques on growth, yield, fruit composition and wine quality of Gewurtztraminer. Am. J. Enol. Vitic. 40: 121-129. Reynolds, A.G. and DA. Wardle. 1989b. Effects of timing and severity of summer hedging on growth, yield, fruit composition, and canopy characteristics of ‘De Chaunac’ l. Canopy characteristics and growth parameters. Amer. J. Enol. Vitic. 40:109-120. Reynolds, A.G. and DA. Wardle. 1989c. Effects of timing and severity of summer hedging on growth, yield,fruit composition, and canopy characteristics of ‘De Chaunac’ 2. Yield and fruit composition. Am. J. Enol. Vitic. 40:299-308. Roublelakis, K.A. and W.M. Kliewer. 1976. Influence of light intensity and growth regulators on fruit set and ovule fertilization in grape cultivars under low temprature conditions. Am. J. Enol. Vitic. 27(4): 163-167. Ruhl,E.H. and PR. Clingeleffer. 1993. Effect of minimal pruning and virus inoculation on the carbohydrate and nitrogen accumulation in Cabernet franc vines. Am. J. Enol. Vitic. 44(1):81-85. Shaulis, N., and R. Smart. 1974. Grapevine canopies: management microclimate and yield responses. In: Proc. XIXth International Hort. Congress. pp. 255-265. ll-18 Sept, Warsaw. Smart, R.E. 1974a. Photosynthesis by grapevine canopies. J. Applied Ecology 1 1:997-1006. 27. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 71 Smart, R.E., N.J. Shaulis, and ER. Lemon. 1982b. The effect of Concord vineyard microclimate on yield. 2. The interrelations between microclimate and yield expression. Am. J. Enol. Vitic. 33:109-116. Smart, R.E. 1984. Some aspects of climate canopy microclimate, vine physiology and wine quality. In: Proc. First Intl. Symp. Cool Climate Vitic. Enol. D.A. Heatherbell, P.B. Lombard, F,W. Bodyfelt, and SF. Price (eds). pp 1-18. June 1984, Eugene, Oregon. Oregon State Univ. Smart, RE. 1985. Principles of grapevine canopy microclimate manipulation with implications for yield and quality. A review. Am. J. Enol. Vitic. 36(3):230—239. Smart, R.E. 1987. Influences of light on composition and quality of grapes. Acta Horticulture. 206:37-47. Smart, R.E., J.K. Dick, I.M. Gravett and BM. Fisher. 1990. Canopy management to improve grape yield and wine quality - principle and practices. S. Afric. J. Enol. Vitic. 1123-17. Stergios, B.G. and GS. Howell. 1977. Effects of defoliation, trellis height, and cropping stress on the cold hardiness of Concord grapevines. Am. J. Enol. Vitic. 28( 1):34—42. Steel, R.G.D. and J.H. Torrie. 1980. Principles and Procedures of Statistics: A Biometric Approach. McGraw-Hill Book Company, New York, NY. Weaver, R.J. 1958. Effect of leaf to shoot ratio on fruit quality and shoot develeprnent in Carignane and Zinfandel wine grapes. Am. Soc. Enol. Vitic. 14:1-12. Weaver, R.J. and RM. Pool. 1968. Effect of various levels of cropping on Vitis vinifera grapevines. Am. J. Enol. Vitic. 19:185-193. Winkler, A.J. 1930. The relation of number of leaves to size and quality of table grapes. Proc. Am. Soc. Hrt. Sci. 27:158-160. Winkler, A.J., J.A. Cook, W.M. Kliewer and LA. Lider. 1981. General Viticulture. University of California Press, Berkeley. Yang, Y.S. and Y. Hori. 1979. Studies on retranslocation of accumulated assimilates in ‘Delaware’ grapevines. I. Retranslocation of 14C-assimilates in the following spring after 14C feeding in summer and autumn. Tohoku J. of Agr. Res. 30(2):43-56. Yang, Y.S., Y. Hori, and R. Ogata. 1980. Studies on retranslocation of accumulated assimilates in ‘Delaware’ grapevines. II. Retranslocation of assimilates accumulated during the previous growing season. Tohoku J. of Agr. Res. 31(2): 109-119. .50... 09.0.". 0.5.53. 302 9:00.50 050... 0..—5.00 55.! 5.3.0000 2.00.2 as 58.58.» .2. .o ”.... a... H... 8 a 88.588 822 n. 0050.2 0002.20.36 m... 0550.0. 05> coca... 0+0 :0 .0. 2803.0 .0 .BE:_. 05 2 9520000 .00 003 .0>0. no.0 000.00 0.... 50.. 03.00. ...05 v a 052:0 0050 00.652 :02... 0.00:0 0530... 0:0... 050: 0023050000 003 208.00.. 02.01 N 60582 0uo£20850 rev 05522 05> 00:20 0+0 :0 .0. 2203.0 .0 .095: 05 9 050.0000 .00 003 .0>0. no.0 05.5000: 9. 50.2 2 00.5.0 203 005., w...2.r ..an. .5 0050.0. 03:20.33 m; 2 0055: .8050 .030: 2.0 0+0 2 coca... 005., .2200 x 0092 0.03 05> 0..: :0 2203.0 .0 $007003 .28 ”$8.84 £8-88 ”4.8580 £863 22.3 8:8. r... a 8.3 8.333 5.2.8.8.. on; .2 2.5 3 .o/. .88. so; 282% o... 55.3 .352. .o «.878»... o8 ”$8.8"... ”$8.88 secromum .....omou. 0.2.3 8:5. m... a 052. 02033.6 5.2.838 on; .2 .0820 55.3 > .. 0: 2. 0c 0.. . 0.. 0c .. 0c . o o.~ on at 8.0 So c 3.. no. 9:. 8 35 on a 3.0 N .258. .... 008: o 3 Nu «.3 So So a mad I. 0.3. a 8.8,... on a «to a .8. use.» :3. a mo c.~ no. on... E.» c 8.. 0o 93 n :3 cu o «3. x 8+0. .280 3 on. > on. 8.8 5.22 In .23 .5. .830 80 .53 .5. 8852 35>.va 3:53. 9.22% 05> .0820 0_n:_om 0.022.... >..0m .00...0m Fem 0.. .2020 .8020 22> 2.0.500... .80.. .F ..00 00.0020... 02:05., >0..0> 5.0“. .0 005.5003 .0250»). .0 000.05 5.020 53.. 05 use 2:050:60 503 05 :0 003 5 E020 05.5.0 .0 00:02.5 .. wam 00:3... 0+0 :0 .0. 0.0.03.0 .0 .0023: 0:.0. 050.0000 .00 00.... .0>0. 00.0 000.00 0:. 20.. 0200. :05 v 0 00.0.30 00:00 00>022 :25... 0.00:0 05000: 0:0: 0503 00:0..020000 00.... 202.00.. 0000... ~ 0050.0. 000520.020 0.. 0550.0. 0:... 00:3... 0+0 :0 .0. 0.0.03.0 .0 .0023: 0:. 0. 050.0000 .00 00; .0>0. 020 0550000: 0.. 50.0. 0. 00:3... 20.... 005> 05.0... =3“. > 0050.0. 000520.020 0.. 0. 00:5:. .0.03.0 .030: 0:0 0+0 0. 00:3... 005.. .2200 x C. ii ... I. C .I CC ”C o o 0 m. a m. 0 0.. a .. a e 0 00. 8.0 N .0200. .... 0000: n 0 n 0.. n 0.. a I 8 o. 8 z n 0.. 00.0 . .0... 0......» =3". 0 00 0 0 0 ..0 0 8 0 0. 0 .0 0 2 00.0 x 6+0. .2250 069.0 0.02.0 .020 :0. 05> 088 .82 850.2. .053... .0353. a 2300202 2300 .00232 .00002 >0. 0~.0 05>. >0. 30 05>. 00002 020 05> 202.00.... 60:0 000..oo:0 5o 000:0>_.0.000> 000:0>..0.000> 000. .000. .. ..00 00.0020... .00.0>05> >0..0> ::0... .0 005.6020 .0....00 .0 000.05 050 05> 0:. :0 000. 5 0202.02. 05:3... .0 00:02.5 .0 w..m<.. 74 .38. 35m @3222 302 2.350 9:»: 2.628 55.3 c3838 «cue—2 .35 285cm.» .2 a 5; ..S 6 «.m a 285:9» 82...; n. .8522 08,5332". m; 9:552 05> toga m+o ca .2 232:0 .o .383: 05 2 9.6803 .3 was .96. no.0 .5200 05 So: «:62 soc. v a 029.5 853 .3389. so...) 23;» 9:30: can: as»: unco._nEoooa an! “co—Eco: 0231 N 60582 32.38620 r... @5582 05> vocaa o+m ca .2 923.0 .0 39:5: 05 2 9.288 .8 on: .0>0. no.6 .oszaouoc mc £82 2 ocean 35> 2.3:. .....u. > .3538 08.5.0330 m; 2 3:55 .0520 .930: new 95 2 3:33 35> .2200 x . . . .. .. : 9. 2. . a not a «8 a 8: a 3 n 9 a n. on cm a no n 328. .3 .58... an 28 no 39 a 83 na 8 na 9. as om S E a 3 > as 2:5 .3“. a 89 a 88 a 9.9 a I a E a mm mm 3 a no x 8+8 .2500 $23 .85 31:3 $25 82m 695 .223 895 32E 32.3 33 553 .23 A953 33 $3.52 3.5.3 302 .2... 602m. .223 .095 baEtm 3230.. 3263 3263 302. 33 .025 .223 .096 EaEta 22:23:. .28 :32 .53 :82 33 .28. .28 :82 33 \ .32 33 .82. .F .60 c2322.. .mEgocS >23) Eon. 8 $5335 .933 .0 32cm... 3 no.8 30. .002» :o mom? 5 6033 9:55 3 00:02.5 . .n m..m<._. 75 .005 09.0.. 0.0222 302 0:00:00 050: 05:28 55.3 5:90:00 0:005. .35 28.590 .2 .o x... 0... ”C .00 .0 58:20.0 0022 n. 0050.0. 03:33.02". 0; 0550.0. 05> 00:2: 0+0 :0 ..o. 0.0.020 .o .0005: 0:. 2 050.0000 .00 m0; _0>0_ :20 .5080 0:. ES. 0200. :05 v 0 02050 00:00 00.650. :25: 0.00:0 05000: 0:0: 050: 00:53:82.0 003 £05.00... 0000... 0 0050.0. 000E2202o 0.. 0550.0. 05> 00:2: 0+0 :0 .0. 20.020 .0 .350: 0:. 0. 050.800 .00 003 _0>0_ no.0 0520000: 00 50.0. o. 00:2: 00:3 0225 =0”. > 0050.0. 000520.020 m; o. 00:55 .8020 .032. 0:0 0+0 0. 00:2: 00:5 .2200 x . 2 2 .. a: . .. . .. a: n 3.0 003 9.0. a 000 0000 a 030 a B n 005 n 00 on N .0208 .... 02.0... .n 0.8 300 00.0. 0 o: 0000 a 0000 n 00 an 2.8 as 8 S > 60. 2:0... =3“. 0 0.00 800 B0. 0 000 :00 a 080 0 ov a 08.. a i «m x 8+0. 6080 8.9 22> .2020 05> .830: 2 .830: .0 coaso> .0 320:0. .0..on820 82.0 .920. .03 Anseflczae 29.50. 10:50. 201:. 3.00.0) ES. 0520300.. 0520300.. .0952 .025 .002 .004 32.00.. .002 .00.. 20.500... 002 .00.. 00.< .00.. 002 .00.. £320 .00.. _0.o.r 3.0.. 05) .aE0m .085 .08... _0.o.r .009 .F .80 00.0020: 00.03:; >0=0> ::0H. .0 00:30:05 .0>>0m .o .0030: .0 00.0 .00_ 05> :o 000? 5 80.03 05:3: .0 00:02.5 .0 M30... 76 ..m «5.21561? "3. 88.0 u 9}. Es 152m... 5530 mm ow 3 S m o _ > p » L . . p _ p p L . p h F p p p p o r com .1 o9. f f o8 . r 0 com 0 w \ o f o r coop 8m. 353:3 >m=a> can". 5 85535 _m>>mw 5 9589335 5.3.0 .3 236: .2 235% 89 .F .9“. HBQWHN HEMO'H Leaf Area (cm‘2) ééé 8888 O 77 PRIMARY SHOOT LEAF AREA a . / r 3 o . o I 0 %PO . . o : 07%. 3 ~/. ; 7° 0 3 e 1 ' .; o 2 0/ .0 1 o 1 3 '- 4 . o E—v-F-FE Y T a Y A 1 a , r fi fi x O 2 6 8 1O 12 14 16 18 Mid-Rib Length (CM) H“2=0.9798 t(x)=(0.7377) ' (X‘1.9793) Fig. 2. Standard for primary shoot leaf area measurements in Seyval grapevines at Fenn Valley vineyards in 1993. 78 LATERAL SHOOT LEAF AREA 180 160 D 140‘ E; § L1LL1|4L llll 1111 11 A Leaf Area (cm‘2) O O A! 8888 L14 [All O I; 2 4 6 8 1O 12 14 16 Mid-Rib Length (cm) 0 R"2=0.9804 f(x)=(0.7591) " (x‘t .9629) Fig. 3. Standard for lateral shoot leaf area measurements in Seyval grapevines at Fenn Valley vineyards. 79 VN .888 9.365 89 9: 9.26 «29823 >m=w> econ. «a $5586 .miow 5 A88 £90. 80% basin :0 5:939:50 .60ch .o 853:... 6 .9“. 595: ..N - hwmnm 03m W.M.—h; mxmm>> or m_. - p com_m.m> N F m 802m C I O N l O V r O Q g 8 (ma) H19N5n Loon-rs Auvmud é mwowI I w_._._wm._. .33“. ..ll JOIPZOO ll o: 80 :83» @5266 89 mg. 9.26 356:3 >o=w> econ. 3 8:32.05 _m>>mm s cozozuoa fie. 605 >853 .8 5293928 .888 .o 853:... .m .9". Pmem 03m Emit. wxmw>> “$me com§m> 802m PNLL ,m.FL_,_m.w.b__WFL...@....m_..Lm.._Lo ¢¢ CC OC To . r an m a mm m H w i d 3% r d - M r W a f 9 .2w 1 H .A ,. m”. a I am ”cum 1.1111 a H a i 1 mm wwowI III mqfimh ...—Du. ll. ..OEPZOO ll. _ 8] .838 0::st 89 $5 05:6 3.98:3 >m=w> econ. La «05585 628m 5 8:885 605 .929 cc cozmszcoo 3058 Lo 00:25:. .o 9.... .513m 03m mwt< wxmmg meEmI com_mLm> Eoo_m rm 3. mw NF m m o o p L L b L L L L L _ L _ L L _ L L F L L o Cfl L... W I N I v w LL 9 we r m L w m r I 0.. m. w H N.. wODMI I w_.jwm._. .33“. ll 405.200 II HSBWHN WVHELV'I .commmm 9:266 82 05 05:6 3.93:3 >e=m> 50¢ E 823399 _m>>mm E cozoanoa Law. 805 3.03. :0 5:93:50 .683 .o 8522. N .9". hmem 03m me< mxmm>> mate: :8 fiLm> £02m vm rm m F m P N F . m m m o > > IP — P p . o L H. or H, m , om a , V T 1. low E Y I: s N H n \ we. own , 3 , H w \ 1| 8 m 00 meow: ll m...._mE.:E I .6528 II _ .633 65266 89 9.: 6566 366:3 >m=m> ccmm 6 352866 .965 5 A53 £266 696 666. :0 6293968 6960 6 coca—55 .m 6E ...mem 03m IMP”? mxmmz, 66361 com_m6> Eoo_m em a 2 3 «L 4 a m m o L L _ L L _ L L P L F 5 L L _ L L _ L L _ L L O L wow 6 we :8 .3 18. m H 8L moowI III. m_._._wm._..._._3n_ ll. JOIPZOO Ill (we) H.LMOUE) .LOOHS WHEN! .688 65266 89 65 6566 666:3 6:05 56.... 6 $5636 613 5 A53 5266 626669 665m .99 05 :0 6:96:60 60:8 6 882.5 .m .mE mezm 03m Eur—,1 wxmm>> meamI com_mLo> 502m 6 6.3 9 «L 6_Lw__m,_o , L3 ,0 m H #« C c 1..I II.— ,om m , V , 1| . A .169 mas L m I H. \\ .62 A \\\ r ...: \\ 9 \\ . w \\\\\ ICON W \,. x r H , m .romwmw So 68: ...... 9465.33... II .65.on :: 85 :80me 65266 89 05 6560 666:3 >0=0> ::0u_ 6 856006 .013 5 60.6 :00 $60. .06. 05 :0 6669600 60:00 6 00:026. .0? .9". ...mem 03m me< wxmmg 60>sz :0066> £02m em a 9 9 «L 0 0 m 0 L _ _ L L _ L L _ L L _ L L _ L L _ L L _ L L 0 mo. L 0. wow I. , V H .I won m woe w r d H 3 ..rom H , S H H ...00 m x 7 H12. m H 00 MODmI II. m_.3mm._.._43u_ III. ..OEPZOO III _ 86 vm .668 65266 89 05 6566 «066:3 6.6) ::0n_ 6 356066 663 5 65> 66 66.606 666 60. :0 :5669600 60:60 6 00:62.5 .2 6E ...wmnm 03m mmt< mxmw>> 60261 :om66> 502m FN mw 9. NF m w m o r L _ L r _ L L _ L L _ L L — L L _ L L _ L L 0 w: I CC . I 7 n H ,rm Nd I? ..l L a mm H , 8 pm a , d H a ”IO—.IAI T N mehm . w 1 V . hcx ,rE. m2 ,2 MODMI I w_._._mm._. .33.... III ..OmPZOO III CHAPTER III THE USE OF CARBOHYDRATE SINK COMPETITION TO REDUCE FRUIT SET AND BOTR YTIS ROT IN SEYVAL GRAPEVINES Abstract Three pruning severities and two times of thinning were applied to Seyval grapevines to vary the amount of fruiting and vegetative sinks at fruit set. Vines were balanced pruned to either 10, 30, or 60 nodes per 454 g. of cane prunings. Half were flower cluster thinned and half post-set thinned to 15 clusters per 454 g. of cane prunings. In 1993, a third factor of cropping level was imposed across each treatment. Post-set thinning decreased fruit set, yield, and Botrytis bunch rot while increasing berry weight and soluble solids. Losses in yield could be overcome by decreasing thinning severity. Increased sink competition created by greater numbers of apical meristems, had no effect on fruit set or Botrytis infection, and decreased cane cold hardiness. Post-set thinning to increase sink competition and reduce fruit set appears to be a viable cultural practice to reduce Botrytis cluster rot. Introduction Botrytis bunch rot is a severe disease in Michigan’s cool, wet climate. Its control to prevent the loss of yield and fruit quality can be a costly endeavor for the vineyard manager. Chemical control and sanitation are successful approaches to minimizing Botrytis infection. However, chemical control is expensive in Michigan’s climate and control products are becoming more limited. Botrytis cinerea lives most of its life as a saprophyte, getting nutrients from dead or dying plant parts (12). Sanitation practices reduce the amount of inoculum in the vineyard before each growing season by removing old clusters, dead wood, and debris from the vineyard. This can be costly and does not completely relieve the problem. 89 Selecting for cultivar resistance and canopy management are also methods to control bunch rot infection. Unfortunately, some cultivars used in wine production are susceptible to infection. The cultivar Seyval (S.V. 5276) was chosen for this study because of its importance in Michigan’s wine industry and serious cluster rot problems. Seyval vines are susceptible to Botrytis infection because they form large, compact clusters. It is for this mason that they are also easily overcropped. Seyval is managed using severe balanced pruning and flower cluster thinning. It is recommended that a 15+10 balance pruning formula with thinning to 17 clusters per 500 grams of cane prunings be employed to Control crop level (17). The potential for yield increase is great if canopy management PTaCtices are developed to control overcropping and help inhibit the infection of Botrytis blmCh rot of Seyval in Michigan. Botrytis cinerea may infect grape flowers and it can remain latent in these tissues Until the clusters begin to mature (13). The spores require prolonged periods of free water and nutrients on the berry surface for germination (12,14). The fungus produces enzymes that can destroy the integrity of the berry in less than 24 hours (12). The main protection for the grape berry to infection is its skin. The cuticle membrane and the epicuticular wax 1)th ide the primary physical barriers to pathogen invasion (1). For beny infection to (bent, the pathogen must find a weakness on the berry surface where it can bypass the shtick membrane, or directly penetrate the surface (3). The epicuticular wax layer 1‘11 uences the retention of pesticides, the water retention of the berry surface, and the E‘CIII‘ésive ability of plant pathogens (1). Percival et al. (15) found that exposed clusters and be having little berry contact, produced more epicuticular wax and cuticle. Clusters with g berry contact may be achieved by reducing the number of berries set per cluster. 6i The competition for carbohydrates and growth hormones between sources and “ks to influence fruit set has been well documented. Intra-vine competition for Nbohydrates and growth hormones by vegetative (22) and fruiting sinks (7) influences \ 9o fruit set. Decreasing the competition between sinks by shoot tipping at bloom (5,6,18), girdling (2,5), or flower-cluster thinning (23,24,225) increase fruit set. Heavy pruning ‘2 (23), leaf removal during bloom (4), and within-cluster competition decrease fruit set (11). ) Flower clusters compete for carbohydrates during bloom (l6) and after anthesis when the shoot apex and the developing clusters become the strongest sinks (6,9,16). Flower- cluster thinning increases pollen genninability (25) and shoot growth (8,11). Post-set thinning can decrease fruit set by increasing the fruiting and vegetative sink competition for assimilate and growth hormone (7,22), thus loosening cluster compactness. Post-set thinning reduces berries per cluster in Seyval, and a compensatory increase in berry size and fruit quality results (10). Vegetative vigor and the development of laterals decreased When thinning was delayed until after fruit set (10). Potentially the number of benies per CIUSter can be reduced, resulting in less berry contact and thus less Botrytis infection. E . 1:]. . 1 To explore the effect of reducing pruning severity and/or post—set cluster thinning to increase competition between sinks through anthesis in Seyval. 2- TO determine if increasing the amount of fruiting sinks at fruit set will decrease the 3 number of berries set per cluster in Seyval. I T6 determine if increasing the amount of vegetative sinks at fruit set will decrease the 4 number of berries set per cluster in Seyval. - ’1‘ Q investigate if decreasing the number of berries set per cluster by post-set cluster 3 thinning will decreases the amount of Botrytis rot in Seyval. \ .1‘ Q determine if the loss in yield in Seyval due to the decrease in fruit set can be overcome by retaining more clusters per vine during thinning. k Materials and Methods ElantLMaten'al The experiment was established in March of 1991, 4.5 miles East of Lake Michigan (42- 15’ latitude) at Fenn Valley vineyards in Fennville Mich. Mature, bearing Seyval grapevines planted in 1975 on an Oshtemo sandy-loarn soil were used in the experiment. Rows were spaced 3.0 meters apart with vines 2.4 meters apart within rows. The vines were trained to a bilateral cordon on the top wire, 1.8 meters high (Hudson River Umbrella training system). We By leaving three levels of nodes retained, three levels of sinks (vegetative and fruitf‘ul) were established for each vine during the dormant season pruning in an effort to i“Crease competition among sinks during bloom. This was accomplished by retaining 10, 30» (Dr 60 nodes (N, 3N, 6N) for every 454 grams of dormant one year cane prunings respectively. The most severely pruned treatment (N), retained nodes on short spurs along the Cordon. The (3N) treatment kept five node canes; and the lightly pruned (6N) treatment retai lied canes of ten to fifteen nodes in length along the cordon. The (N) vines were pruned to 45-nodes and the live one-year cane prunings ei ghed to estimate the previous year’s vine size. The node number for the (N) vines was ““311 set for the duration of the growing season following the procedures stated above. EQQ~Eause of the number of nodes needed to obtain the (3N) and (6N) levels were much gr'éaser than 45 nodes, a method was devised to estimate the weight of retained wood. ‘\ f‘ . l ‘ {Qt pruning the (3N) and (6N) treatments to 90 and 180 nodes per Vine respectively, the l \r '5 canes were weighed. The remaining live nodes of canes on the vine were counted. A 91 92 weight was assigned to the number of nodes over 45 retained on the vine, based on a node weight calculated in 1992 (113 gm per 52 nodes) and 1993 (113 gm per 50 nodes). This weight and the weight of cane prunings were added, producing the vine size for each (3N) and (6N) vine. Crop thinning is also required in Seyval for optimal crop control. Flower-cluster thinning (PCT) and post-set thinning (PST) were imposed to compare the influence of increased cluster numbers per vine during fruit-set. Cluster thinning was conducted just prior to bloom in the (FCT) vines, and when the berries were 3-5 mm in diameter (23 Weeks post-bloom) in the (PST) vines. The number of clusters per vine were proportioned by using the weight of live one-year-old canes at pruning (vine size). Thus, crop level was balanced among vines based on their vegetative production of the previous year. In 1992, the treatment vines were cluster thinned to 15 clusters per 454 grams of domlant one-year cane prunings. In 1993, three cropping levels were employed in an fitterIlpt to offset any loss in yield due to a decrease in fruit set The (C15) treatment maj litained the cropping level imposed the previous year (15 clusters per 454 grams of dotTrh'c‘tnt one-year cane prunings). Higher crop levels, (C18) and (C21), increased the nu“) ber of clusters per vine after thinning to 18 and 21 per 454 grams of dormant one-year e prunings respectively. § The effect of increased sink competition during bloom on vegetative growth was beined. A consistent determination of vine size was measured as described above. At ‘1 q burst, nodes that did not produce a shoot (shootless nodes) were counted. Percent Shsg . . . . tless nodes, vegetativeness (vrne srze (gm) per shoot), count shoots (nodes retained at \3 . 17hr“ ng minus blind nodes), and non-count shoots (total shoots minus count shoots) were 93 Calculated for each vine. Treatment comparisons and interactions of pruning data were analyzed each year and over both years. E . : l l ' . Point quadrat assessments were taken at three times during the growing season each year. A thin metal rod was inserted into the canopy to determine the number of leaves above the fruiting zone (19). Five insertions along the cordon were made on each treatment Vine at bloom, veraison, and harvest. Insertions were always made from the West side of the canopy. A frequency distribution of the number of contacts and a seasonal history of leaf layer formation were developed. WW During the 1993 growing season, detailed shoot and leaf measurements were taken on e\’ery (C 15) vine of the sub-sub—plots in each replication. Within each sample vine, five Shoots were randomly flagged along the cordon for vegetative measurements during the grovVing season. These includect 1) primary shoot length (PSL); 2) primary shoot leaf number (p315); 3) lateral number (L); 4) lateral length (LL); and 5) lateral leaf number 12.51% ; measured on May 26 (WK3 weeks after bud burst), June 2 (WK4), June 9 (WKS)' e 16 (WK6), June 30 (WK8), July 29 (WK12), and September 23 (WK21) one week mgr 9% to harvest. Additional calculated data included total leaves per shoot (T SLF= “I L’RrLLF) and total vegetative shoot length (PSL+LL). Treatment comparisons were m at appropriate dates and over dates on those variables measured or calculated. Vv‘ At thinning, none of the sample shoots were defruited so that shoot development lth a fruiting sink could be studied for the remainder of the growing season. A count of “a QQ‘s (CS) longer than 10 cm and bearing three or more leaves was taken two weeks prior 94 to harvest. Primary leaves per vine (PSLF*SC), lateral leaves per vine (LLP"SC), and total leaves per vine (TSLP‘SC) were then calculated and compared between treatments. COldHardinessfixaluations Prior to pruning in 1993-1994, a cold hardiness evaluation was taken of the previous season’s vegetative growth. Previous work with Seyval (21) and a re-evaluation Of that work (Appendix 1) has led to a definition of Seyval cane characteristics closely associated with cold hardiness. Classifications were created from hardiness analyses done on Dec 19, 1991 and Jan 22, 1992 on Seyval canes (Appendix 1). Periderm and primary bUd 'TSO’s were determined for variations within four cane characteristics, and over node POSitions along the cane. Canes having a medium diameter (7-10 mm), medium intemode fen 8th (6—8 cm), and dark brown periderm were found to have superior cane and primary bud Cold hardiness. Persistent laterals and node position (2-9) did not have any influence on <=<>1 60.666666 666506: 0.00 6005 0550.56 coon. L0 .6.066600< 66:60me 300660 65 55.3 65> .66 66:60 6 60:6: 65. x 6: 0: m: 0: 0: 6: 6: 6: 6: 6: mm an w? m 0 mo mr adv Fm mmd «omfiom mm mm ml 5 0 mm 0? . 6.: on nmd 5.005 26 we 6* or 0.. ..w mm or 5.6? 06 mmd ~6m60n. 60 6.. mw m 0.. an 2. mm? on omd £006 20 IV 3 t or 06 we mL 6.6? 3 66.0 66600 mm mw m_. m mm : 0r New VF ..md Eoo.m Z 65.... 65:55 x 065666 66002 m: m: . . m: 6: 0: m: 6: 6: 6: on mm or o : mm 6.. 50.. we mmd ~ Swamoa 6.. em 3 6 I 6 3 6.6. no 66.0 L. E026 m:.::.._F x .6650 6 655 . . . . 6: . . . . . . . . . . . m: 6 mm 6 60 hr am 00 6 mo 66 3 n «.3 600 mmd 20 0 we n 9 t. 6m n C n vo n or n6 6.9 0 mm vmd Zn n 06 n m_. we 6 m 6 mm 0 C. 6 or 6 new; 0 6.. mmd Z .0L+0Luz. 6.056... 65:05 6 06:66.". 66002 06:60 x 06:60 x 66:60 x 66:60 060:6 060:6 66002 900:6. Em 065661 5652me 66566.... .665 3:600 5:600 3:600 6:00:02 6:00 0:..m ox. 6~.6 6:.3 66002 3% 65> .006 6.066600< 5:6..60xm 60265660; .66.... 6...... .5“. .5 5.6.53 .55 ... we. 6%. ...e. 866 «8...... ON.m 2.... 0... :0 8.5.... 6.3.0 .0 65. 0:6 5.2.3 0555. .o 862.... .L 6.65 104 .566... 66:66 6.05.0.2 362 6.:60:0o 6560 6500.00 55.3 606.6066 6:665. .65 58:00.6 .0: .0 .... 0... .... 006 .6 58.....06 6626., H. .:0.666. 6. 6:50.58 660.0 60 66:60 6 666660.60 65. N .:0.66.:. 6. 0: 6:50.56 6060.0 .60 66:60 6 .6080: 65. .. 0.00.: o. .0.... .60. :66. 6.6.60.0 .6266 06 6 6665666660.. 0:6 66000 .00.. 606.: 0606.6 6 :0 06660 6.6050: 630.“. 066.503 x 6: . . . 2 .. 6: 6: 06.6 0 s 0 600 6 ..N 0 v: 666 6.6 .66 606. 66.6 6 66 6 060 6 6m 6 6'6 666 6.00 E85 :6 66.6 0 65 0:6 66w 0 ...N n 66.. 606 0.0.. «66 60a. 60.6 6 :V :6 66w 0 nv 6 606 006 0.: 500.0 26 06.6 0 6m 00 66.. 0 66 a 66.. 606 6.00 .66 «won. 669 6 ..v :6 63 0666 6 066 606 6.60 802m 2 6.5... 65:55: x 06566.". 66002 . . . . . . . . . . 6: 6: 00.6 60 me mm 060 606 6.00 06m «606. 60.0 0.. 06. 06 600 600 0.: 506.6 65:55. 660.0 6 65.5 . . . . 6: . . 6: . 6: 6 66.6 0 mm 660 n 06 696 n 606 6.05 26 6 06.6 0 06 060 n 06 666 6 666 6.: 26 n 60.6 6 66 060 6 3 066 a 606 6.00 2 .0F+0.uz. 6.056“. 05:26 >0 065666 66002 .50. ~ 6.60.0 > 6060.0 660.0 66:566.... .53 ....66 .66 2.66 66:66 66 ....66 .66 66:66 x 600.02 ...:0. 56:60 66066 .5600 0:60.66 .66 ... 630.". .66 6560.0 00...... ...... ...... 0 0......5 ...... .. 60...... 0.0.. 0000.06 .00 ...... ..0 00.00.... 6.00.... .0 0..... 06 5.0.00 00.0.00 .0 00:02.... N 36.: .52. oucam 03:32 302 $59.30 9.8: 255.8 55.3 5:233 mama—2 .35 235:9” .2 6 EL .x; ”C .xb 8 E825» 829 n. .3555 2.3 938520 9 2 2:95am: 6w Eon 05 E .3 an; .80. no.6 N .mmEcEa 980 233520 92 2:953: 802m 9.: 5 «cm was too. no.6 > 105 m: w: I . . we we 8.3 m6» 0 Dim 0 EN 0 ova. Pm and 8m Eon no.3 v.03 a 3.2. a one a ops an 3.0 £02m cm no.3 ad: a 9.3 on 8N. on go ov amd “ow 50m v0.3 mdow m N92. nu «Na an 56 mm omd Eoo_m 2m of? fin? on RN; on Ed on 51m er and «cm Eon 053 «$2 n «0.3 on vow. on cam «F 5.0 6006 z 2:; SEE x 8522". «302 . . . . . . . . . . w: m: 3.: “mow 3.: and owe me mmd N «om Eon 3.9 02: 2.9 New Ed 3. $6 > 585 $3:th 5.3.0 .o E? mc I m: we m: .. w: 3.9 a 0m? 3.? med mob m cm omd 20 m9? a No? 5.9 and mm.m n mm 3.0 Zn 8.9 a 0.09 3.3 3.0 mm.m o 3 mmd z Aow+owuzv Sacco”. @555 3 conflom 882 35m 9.3 ; 9025 85m 05> S AS .5 35%me uoESom 35>axv EoEEcC. 225322: >53 22: 529.223 22> 22> 8.82 £5 25 33 no.0 30:322.... 33:03.35 as gig; g; g: a “2.30% 3% s 2% .§ 5 bag: as so: as 0% 8.; 2. S Page .230 3 as... can €28 Bean 3 8:255 .m 39: .mULm>®C_> >m.:G.> CCGU on. l1!" ‘Iilf- Amxv Em!) C0 0C.CC.£u «w 00...:Qu0L m_0>0_ QOLO msmwmgoc. .LO 00:01.1“:le 106 .89 .m >3... no 355: 5620 925 85> 2.an on... N .89 .3 0:2, co 8:55 .236 So; moo; 2.0“: on... > .mmcEEn 9.8 .o 2:90. Qmmv .on 992:0 E no .9 .3 2 can: «omnmon .6 Con: 88.2 556 an $58.23 _m>o_ no.0 x Kim mmk Fm mo.m mmd m? vm.m 56 9 2m vmh Sim Fm 5+ mvé mp 36 3h 9 2m mm.m wm.m Pm mfm 0N6 m— mmd mvé m_. z N x x .05.. mEmmoG 5n .5“. >2 >Eo>mm mcEEn .wEgocS >o=m> econ .m 855nt .mgow no mm? s as 2m; 8 9:22. a 8:52 «Eon use 8332 3 Eggs .V «BE O UGO—-NQGF E0: UGOGNO)! ,OUNG>0C.> >0:G> CCOK .5 QQCQIQQQNU .I))°m x >2L0)00 UC‘CULQ ‘0 00:03:93 m WdexF loony—Ur: )a._vu3nv :3: RUE-w UnCUCOQEOU U.c.> fir: CO UC.CC.C~ L0~fi3~0 no AVE: DEG 0th 35¢ 29:22 302 95250 053 2.628 55? cozaanom 2.3.2 .65 anzegm .8 .o 5.0 x; .8 gm 8 Eagzcoa 83.9 ... .3555 050 ...—3.23.0 m? 2 2:25am: «om Eon 9: c_ «om was too. no.0 N .oucEEn ocmo 0.6.23.0 m? 2 2:958: £020 05 5 wow mm; 32 no.0 > m: .. m: we we a: . o... . N.m a 0.0? 05.0 0N0 0 omN n fiNwN NN o ové “ow «won vN o 0.: who 0N0 N NN.N a 0.000 0N o ohm E86 20 N.N 0 0.9 050 N0 F mvN n meN 0N on 00.0 now «won 0.N on 0.: ~50 :0 N 00.N . a 0.m0N 0N an 50.0 8005 20 we n «.3 N50 0N0 N NNN n 0.NNN VN on No «00 «won 0... n 0.0? 2.0 NM 0 00.? n 0.0VN VN on 004m 6005 2 25.. 355 x 02.38... moooz . .2. w: .. S. .. .. mc .. m vN 0.2. 05.0 mN.m F 3N méNN MN 006 N wow Son 1 rm 0: to a; w 8N PEN mm (no > Emgm 9.2.5:» .0520 .o oE:. .. m: w: m: wc .. . m: m: m 0N ...mw 05.0 0N0 m a omN no ¢.0mN mN 00.0 20 o ..N 0.: 050 3.0 N o NNN m 0.00N 0N 00.0 20 n 09 N0? v50 0N0 N 0 SN 0 EmoN VN 00.0 2 3:3qu 235.0“. ascan B 8:32“. 882 .236 mozom €23. In 223.0 3.9 2:9 .3532 $529: E953; \Boocm o_o:_ow 235:: Ezooéoz ..3 >..om .E .6620 .0520 En; .mo.m>oc_> >o=a> :5... 6 3553.0 i>>om .o 000..-ng 89: 300.05 3205 3:95 :2. oco 2coconEoo 29> 9.: co 05:55 5820 .o 0:.: 0:0 3:93 chEn .o 8:255 .m m...moc_> >m=m> econ E «83890 513 E $5893me .296 .on 2950.. .o. Emocflw Nmmp .P .9". H 00N , 00¢ H 08 M a. H as com? HEGWHN HEMO‘L-J 109 ##m Emotéo 33% "3. 886 u «E Ea 1525 $530 mm om m. 2 m . o O O \ 1 000 .mo.m>oc_> >o=m> econ E «83390 _m>>om 5 25893988 .938 .on 22.6.. .2 059.30 80. .N .9... HBQWDN HBMOWd 110 $95: com_m.m> Eoo_m 20 ll 20" 2| JeqwnN JSAB'I JBG‘I .882. .99 32 mm Emmenxm .wo.m>oc_> >o..m> econ. E «83890 aimm .0 593292. .8058 .888 Nmmw 9.. co 3:98 0555 .o 82m .0 .9”. 111 59.9.. com_m.m> Eoo_m 20] 20' 2| .fiimm .o .882. .98. .mm. mm commenxm 3.98:3 >o=m> cam“. um moc_>onm.m 3280 .0 2.952920 >nocmo 83$ mm? 9.. :0 3:93 mEcEn .0 Spam .v .9”. , no m... mN JaqwnN JaAe'] men :8000 05.590 000? 0:. 05.30 00.30:; 0.3) ::0u. .0 023890 .0100 .o 20.0.0. + :0E_.nv 62.0 .0n 50:0. 02.0.00? .08. 0:. :o 3:050 0:.:=.n .o 00:02.2. .0 .0.". ...mmDm 03m mwnhz mxmmg «00201 :om_0.0> E005 NN ON 9 0r 3 NF 0? 0 0 v N 0 . _ _ _ . . _ . . _ r n L _ . _ . _ _ _ . 0 CC a. . 100 I. , m V D ..l r w loop m W... 20 o N 2.. , .I row; m T 9 18N( . z m T omN 113 :8000 05.590 80. 0:. 0:..30 mo.0>0:.> 0:0) ::0... .0 00:.>0n0.0 .0100 .o 20.0.0. + 00.5.3 .85 .0n 0900. .06. 0:. :o 3.050 0:.:=.n .o 00:02.:. .0 .0.“. ...mmDm 03m th< wxmmg .00>.0I 50.0.0) 836 NN ON 0 w 0 w E. N _. 0 _._ 0 0 V N 0 p . . b r _ . P F _ . p » — . P p _ p _ . fifi 1.“ . 20 o m T 20 0 , W Z 0 n 114 NN .:o0000 0:590 000. 0:. 05.20 00.00:; >0._0> ::0u. .0 033800 .0260 .o 69.0 .0n .0052: 0.0.0. :o >._.0>00 0555 .o 00:02.:. N .0.... meDm ODm EMF”? mxwm>> .0020: :o0_0.0> Eoo.m ON 0.. 0.. 3. N _. or 0 N 0 . L _ p p . p . p . p, . _ b O IN I V 10 I m I or 2 f HSGWON 'IVHSLV'I 115 .m 08.0. - 3 .03 n8. 000.00 "NE 0.3.0340 mun. kmm memwm 00N 00.. _ g _ L _ . 0 0mm 00m 0mm 00.. cm _._,...+___ ..F....__r+ — ~ _ - ITfTYf l0 0 O ,_ to f. IO N .LOH SLLAHiOB lNEOHSd O (‘0 ID (0 ITTTIIiTiTIIIIIIIYTYYTrrTTfiIYYITTijTII ID V’ .00:.>0n0.0 _0>>0w :. .0.020 .0n .00 00:00 .0 .0063: 0:. .o 5:05. 0 00 .o. 0...»..om 000.:00.0n N00. .0 0..... 116 ...... 000. I - 3 00 ..o no... 0.00.0 "mm mwkmzqo awn. ._.mw mmEmmm omm 000 omN 00N om? .oow cm 0 ...... .. ........ .......... L p I fir l T Y Y O P ,.. 8 Y f T T O . , -. . . , 8 8 103 SLLAELLOQ maouad O ,,. 8 00 08380.0 .900 :_ .0.020 .0n .00 02:00 .0 .0062: 0:. .o 5.8:... 0 00 .0. 0.5.3 000.:00.0n 000. .0 .0.... CHAPTER IV THE USE OF CARBOHYDRATE SINK COMPETITION TO REDUCE FRUIT SET AND BOT RYTIS ROT IN VIGNOLES GRAPEVINES 117 Abstract Three dormant season pruning seventies and an unpruned (UP) treatment were imposed on Vignoles grapevines to vary fruiting and vegetative sinks during fruit set. Vines were balanced pruned to either 15, 30, or 60 nodes per 454 g. of cane prunings. The treatments leaving additional nodes per vine during the dormant season, were balanced pruned after anthesis. Five and ten node canes were compared at each pruning severity. Increasing the competition for carbohydrates between fruitful and vegetative sinks during anthesis by decreasing pruning severity during the dormant season had little affect on f ruit- set. The (UP) treatment decreased yield, due to reduced cluster formation per shoot. The clusters that did develop had fewer flowers each year and thus, fewer berries set. Reductions in Botrytis infection were observed only in 1993 in the (UP) treatment Cane length had no affect on fruit-set or Botrytis infection. Competition induced by fruitful sinks was limited because of the small size of Vignoles clusters. However, increasing carbohydrate competition between fruitful and vegetative sinks through anthesis was detrimental to fruit and vine maturity. Post-set pruning does not appear to be a viable cultural practice for management of Vignoles grapevines in Michigan’s climate. Introduction Vignoles is an important cultivar in Michigan’s wine industry. It bears small, compact clusters that are susceptible to harvest season Botrytis bunch rot infection. Because of the smaller clusters, balance pruning alone is sufficient for crop control. A 15+15 pruning formula on a High Cordon training system is recommended (11). Botrytis bunch rot is a severe disease in Michigan’s cool, wet climate. Its prevention or control can be expensive, and if not sufficient, produce losses of yield and fruit quality. Chemical 118 119 control, sanitation, and canopy management are all successful approaches to minimizing Botrytis infection (14). However, chemical control is extensive in Michigan’s climate and control products are becoming more limited. Botrytis cinerea lives most of its life as a saprophyte, getting nutrients from dead or dying plant parts (14). Sanitation practices reduce the amount of inoculum in the vineyard before each growing season by removing old clusters, dead wood, and debris from the vineyard (14). This can be costly and will not completely relieve the problem. Canopy management practices for Vignoles grapevines need to be developed to help inhibit the infection of Botrytis bunch rot in Michigan. Botrytis cinerea may infect grape flowers and remain latent in these tissues until the clusters begin to mature (15). The spores require prolonged periods of free water and nutrients on the berry surface for germination (14,16). The fungus produces enzymes that can destroy the integrity of the ben'y in less than 24 hours (14). The main protection for the grape berry to infection is its skin. The cuticle membrane and the epicuticular wax provide the primary physical barriers to pathogen invasion (1). For berry infection to occur, the pathogen must find a weakness on the berry surface where it can bypass the cuticle membrane, or directly penetrate the surface (4). The epicuticular wax layer influences the retention of pesticides, the water retention of the berry surface, and the adhesive ability of plant pathogens (1). Percival et al. (17) found that exposed clusters and those having little berry contact, produced more epicuticular wax and cuticle. Clusters with less beny contact may be achieved by reducing the number of berries set per cluster. The competition for carbohydrates and growth hormones between sources and sinks to influence fruit set has been well documented. lntra-vine competition for carbohydrates and growth hormones by vegetative (25) and fruiting sinks (8) int] uences fruit set. Decreasing the competition between sinks by shoot tipping at bloom (6,7,22), girdling (3,6), or flower-cluster thinning (26,27) increases fruit set. Within-cluster competition decreases fruit set (13), while severe pruning (26), or leaf removal during 120 bloom decreases fruit set (5). Flower clusters compete for carbohydrates during bloom (18) and after anthesis when the shoot apex and the developing clusters become the strongest sinks (7,9,18). By increasing the competition for carbohydrates during bloom, potentially the number of berries per cluster can be reduced, resulting in less berry contact and thus less Borrytis infection. The following eXperiment was designed to test this hypothesis and to answer several other questions: E . l :1 . . 1. Does reduced pruning severity and post-set pruning increase competition among sinks through anthesis in Vi gnoles? 2. Do the number of vegetative and fruiting sinks at fruit set reduce the number of berries set per cluster in Vi gnoles? 3. Does reduction in number of berries set per cluster also reduce Botrytis rot in Vi gnoles? * 4. Do these pruning strategies influence shoot development or bud and cane cold hardiness of Vignoles? Materials and Methods HamMaten'al The experiment was established in March of 1991, 4.5 miles East of Lake Michigan (42-15’ latitude) at Fenn Valley vineyards in Fennville Mi. Mature, bearing Vignoles grapevines planted in 1976 on an Oshtemo sandy-loam soil were used in the experiment. Rows were spaced 3.0 meters apart with vines 3.7 meters apart within rows. The vines were trained to a bilateral cordon on the top wire, 1.8 meters high (Hudson River Umbrella training system). 121 Experimentall'reatmems The experimental treatments compared four levels of pruning severity in an effort to increase competition between sinks (vegetative and fruitful) during bloom. This was accomplished by retaining 15, 30, or 60 nodes retained (N, 2N, 4N) for every 454 grams of dormant one year cane prunings respectively, and an unpruned treatment (UP). The (2N), (4N), and (UP) treatments were pruned to the (N) level of nodes when the berries were 3-5 cm in diameter (23 weeks post-bloom). Two cane lengths (5 or 10 nodes) were compared to study the influence of node position within the canopy. One treatment retained short (5 nodes), while the other retained long (10 nodes) canes, after dormant season and post-set prunings. The unpruned treatments were pruned to either short or long canes after fruit-set. The treatments provided increasing levels of vegetative and fruitful sinks during anthesis. The (N) vines were pruned to 65-nodes and the live one-year cane prunings weighed to estimate the previous year’s vine size. The node number for the (N) vines was then set for the duration of the growing season following the procedures stated above. Because of the number of nodes needed to obtain the (2N) and (4N) levels were much greater than 65-nodes, a method was devised to estimate the weight of retained wood. After pruning the (2N) and (4N) treatments to 90 and 180 nodes per vine respectively, the live canes were weighed. The remaining live nodes of canes on the vine were counted A weight was assigned to the number of nodes over 65 retained on the vine, based on a node weight calculated in 1992 (113 gm per 41 nodes) and 1993 (113 gm per 43 nodes). This weight and the weight of cane prunings were added producing the vine size estimate for each (2N) and (4N) vine. The total number of live nodes above 65 were counted on each (UP) vine, and a vine size calculated using the average node weight of that year. 122 iegetatixeMeasuremems The effect of increased sink competition during bloom on vegetative growth was examined. A consistent determination of vine size was measured as described above. Vegetativeness (vine size (gm) per shoot and node) was calculated for each vine and a shoot number taken at verasion in 1993. Treatment comparisons and interactions of pruning data were analyzed each year. During the 1993 growing season, detailed shoot and leaf measurements were taken on one vine in each treatment replication. Within these sample vines, frve shoots were randomly flagged along the cordon for vegetative measurements during the growing season. These include: 1) primary shoot length (PSL); 2) primary shoot leaf number (PSLF); 3) lateral number (L); 4) lateral length (LL); and 5) lateral leaf number (LLF); measured on May 26 (WK3 weeks after bud burst), June 2 (WK4), June 9 (WK5), June 16 (WK6), June 30 (WK8), July 29 (WK12), and September 23 (WK21) one week prior to harvest. Additional calculated data included total leaves per shoot (TSLFzPSLF+LLF) and total vegetative shoot length (PSL+LL). A count of shoots (CS) longer than 10 cm and bearing three or more leaves was taken on WK19. Treatment comparisons were made at appropriate dates and over dates on those variables measured or calculated. ColdflardinessEvaluations Prior to pruning in 1993—1994, a cold hardiness evaluation was taken of the previous season’s vegetative growth. Total one year old canes were counted and each cane given an excellent, acceptable, or poor classification. This approach has been used previously by Streigler and Howell (24). Excellent eanes consisted of ten or more live nodes with the optimal characteristics of medium diameter (7- 10 mm), medium intemode length (6—8 cm), and dark periderm. Acceptable eanes had five or more living nodes but failed in one of the optimal characteristics. Canes with less than five live nodes or failing at 123 least two optimal characteristics were placed in the poor cold hardiness category. Samples of Vi gnoles canes within the three classifications were collected in December, January, and March after the 1991 and 1992 growing seasons. Using the three classifications was a rapid and effective way to evaluate the relative cold hardiness of each treatment vine. E'-S IE ‘EEI' A standard curve was developed to allow rapid determination of flower number per cluster based on rachis length for both years. Thirty clusters were randomly sampled from guard vines just prior to bloom. Cluster length was calculated by adding the length of rachis from the lowest basal arm to the tip and the length of the lowest basal arm. Flowers of each sample cluster were counted and a strong correlation was found each year (Figs. 1 and 2). The rachis length of clusters on sample shoots within the sub-plots were measured and their estimated flower number calculated. At harvest, berry pedicles of sample shoot clusters were counted and the percentage of berries set analyzed The number of whole berries showing no incidence of Botrytis rot infection were counted for rot calculations. Subtracting these quality benies from the number of pedicles per cluster provided a number of the benies infected with rot. Fruit-set and Botrytis rot comparisons were made each year. 11 El lE':l'! Treatment vines were harvested each year within ten days of October 1. Each vine was hand harvested with the cluster number noted and the fruit placed into an individual bin. Each bin was then weighed in the field and a cluster weight calculated. Clusters from sample shoots of sub-plot vines were harvested separately, weighed, and placed into individual bags for berry counts and fruit quality determination. Berries were randomly taken from these sample clusters to create a 100 beny sample for every sub-plot treatment 124 in each replication. From these samples pH, titratable acidity, soluble solids, and beny weights were determined and the number of berries per cluster calculated. Treatment comparisons of harvest data were analyzed for each year. Using the harvest and pruning data, shoots per cluster (total shoots/ harvest cluster number), productivity (yield per previous year’s vine size), f mitfulness (yield (gm) per shoot), and crop load (yield per current season’s vine size) were calculated for every treatment vine and analyzed for each year. E' IE' 11' The plots were established in March of 1991 at Penn Valley vineyards in Fennville Mi. A split-plot design was used with the main-plot consisting of four treatments (N, 2N, 4N, and UP) of four vines each, replicated five times. One guard vine separated each main- plot treatment. Each main-plot was divided into two sub-plots (Short and Long canes) consisting of two vines. Comparisons and interactions between treatments were made each year using the MSTAT-C statistical computer package. Analysis was by ANOVA, with mean separations calculated using Dunean’s Multiple Range Test (23). The f ruit-set standards were calculated using the Delta Graph computer package. The experiment was terminated after pruning in March of 1994. Results and Discussion If S' E Decreasing pruning severity increased the number of nodes retained, and increased the amount of vegetation removed at post-set pruning. No significant differences were found in vine size from 1991 among the treatments (Table 1). In 1992, vine size in the (UP) treatment decreased dramatieally to 180 grams per vine compared to the other 125 treatments of 760 grams or more (Table 1). The (UP) vines averaged over 400 nodes retained through anthesis. The (4N), (2N), and (N) treatments were significantly lower at 130, 56, 32 nodes retained, respectively. Thus, increasing amounts of vegetation (vine size) was pruned away at post-set as pruning severity decreased. Nearly all of shoot production arose from count positions in each of the treatments. Shoot numbers per vine were similar to the number of nodes retained at post-set pruning (Table 1). In 1993, the (4N) and (UP) vine sizes were significantly different from (N) (Table 2). The (UP) vines improved their vine size to nearly equal that of (2N) and (4N), due to many of the shoots being non-fruitful. The (4N) and (UP) vines had the same number of nodes retained (103), 2-3 times more than (N) and (2N) (31 and 59 nodes retained respectively). Shoot numbers decreased with the reduced pruning severity, and each treatment averaged 12-14 adventitious (arising from non-count positions) shoots (Table 2). Vine size decreased in all of the treatments from 1991, however, decreasing pruning severity through anthesis increased the rate of vine size reduction. (hueMatnrity Cane maturity was examined by calculating vegetativeness per node and shoot, and by cold hardiness evaluations. Since nodes retained after post-set pruning and shoot numbers among treatments in 1992, the vegetative response per node and shoot followed the vine size response (Table 1). Because of the low vine size of the (UP) vines in 1992, less nodes were retained during the post-set pruning in 1993. As a result, vegetativeness per node was significantly greater (Table 2). However, no variance was observed in vegetativeness per shoot although, shoot number decreased with decreasing pruning severity (Table 2). Samples of Vignoles eanes within the three classifieations (excellent, acceptable, and poor) were collected in December, January, and March after the 1991 and 1992 126 growing seasons. Periderm and primary bud cold hardiness were significantly different among treatments at the 0.01 level (Figs. 3 and 4). Excellent canes were superior while poor canes were inferior to acceptable canes over years and at each date evaluated. Using these classifications, each treatment vine cane was evaluated. In 1992, decreasing pruning severity reduced the number of excellent quality eanes, while increasing poor quality eanes per vine (Table 1). Decreasing pruning severity to the (4N) or (UP) level, reduced the number of excellent quality cold hardy canes by 50 percent in 1993, but was found not to be significant (Table 2). However, (4N) and (UP) vines matured signifieantly less acceptable quality canes, and (4N) vines produced more poor quality eanes than the other treatments. No significant differences were found in any of the vegetative indices measured for either year between cane lengths. Reducing the severity of dormant season pruning and increasing the amount of post-set pruning, decreased the amount of vegetation the vine was able to utilize and mature after anthesis. ShQQtAssessmems Detailed shoot measurements were taken during the 1993 growing season from five sample shoots on one vine of each treatment replication. Treatments did not vary in primary shoot length or leaf number until after bloom, when (N) shoots were significantly longer and had more leaves than (UP) shoots (Figs. 5 and 6). At harvest (4N) and (UP) shoots were shorter than (N) shoots, and (UP) shoots still had fewer leaves. No differences were seen in lateral number, lateral leaf number, and lateral length until after bloom (Fig. 7, 8, and 9). Lateral growth decreased as pruning severity was reduced. The (N) shoots produced superior numbers of laterals with more leaves and length than any other treatment. The (2N) shoots produced greater numbers of laterals with more leaves and length than (4N) and (UP) shoots. Lateral growth contributed to the separation of treatments in total shoot length and leaf number. Again, (N) shoots were superior to the Ill .IIIIII'I‘IIII 127 other treatments and the (2N) shoots produced more than (4N) and (UP) shoots (Figs. 10 and 11). No differences in shoot growth were observed among cane lengths for any of the variables measured. Overall, decreasing pruning severity and increasing sink competition, reduced the vegetative production per shoot. This agrees with previous work on increased competition resulting from greater crop levels (2,23). ”“11. ”3.: l' : Lower yields and fruit quality resulted from decreased pruning severities. In 1992, the (UP) vines yielded less than half the crop per vine (more than 1.5 tons per acre) than the other treatments (Table 3). This was the result of reduced berry and cluster weights. Titratable acidity and soluble solids were significantly lower and pH was higher in berries of (UP) vines. The (UP) vines produced 70 to 80 percent less crop than the other treatments in 1993 (Table 4). They managed a crop of only a third of a ton per acre, compared to yields of 1.25 to 1.90 tons per acre in the other treatments. This was a result of non-fruitful shoots, resulting in a f our-fold increase in the number of shoots per cluster. The (4N) vines produced 80 percent less crop than (N) vines. All of the treatments subjected to post-set pruning had a decrease in berry weight each year, and the (UP) treatment had a reduction in berries per cluster in 1993 (Tables 3 and 4). However, only the cluster weight of the (UP) treatment was affected. This coincides with previous work with increasing sink competition (10,19). Berries in the (UP) treatment were significantly lower in titratable acidin with a higher pH than the other treatments in 1992 (Table 5). In 1993, the (4N) and (UP) vines produced berries with the highest pH and lowest titratable acidity (Table 6); again, the result of a decreased crop load. However, the (4N) and (UP) benies had the lowest soluble solids each year. This contradicts previous work (20,21) suggesting canopy reductions improve eanopy microclimate. Post-set pruning reduced the leaf area available for fruit and wood ripening in the (4N) and (UP) vines. Increasing sink 128 competition through anthesis caused reductions in yield, as a function of poor berry development, and fruit quality. Due to the decrease in yield in 1992, productivity and fruitfulness declined in the (UP) vines, but a severe reduction in vines size caused crop load to be over twice as high as the other treatments (Table 3). Long canes increased fruitfulness and productivity in 1992. Fruitfulness was significantly lower again in the (UP) vines in 1993, and their crop load averaged 75 percent less than the other treatments (Table. 4). Long canes were more fruitful in 1993, due to a significant reduction in the number of shoots per cluster. Cane length had no other influence on the yield indices of Vignoles. Delaying pruning until after anthesis, reduced the number of flowers produced per cluster in each year. The treatments did not influence berry-set or Botrytis infection in 1992 (Table 5), but fruit-set and bunch rot in (UP) vines Was lower in 1993 (Table 6). Twenty to 30 fewer berries were set by the (UP) vines and half the rot of (N) was observed. The (UP) vines still ripened 30 percent less quality berries in 1993. Long eanes produced more flowers per cluster in 1992 (Table 5), but no variance was seen between eane lengths in any other fruit-set or bunch rot measurement in that year or the next. Longer eanes were more productive and fruitful because they were better exposed and produced more numbers of shoots per cluster. Only at the (UP) level of nodes retained through anthesis, did increasing carbohydrate competition influence fruit-set or Botrytis infection. Conclusions Increasing the competition for earbohydrates among fruitful and vegetative sinks during anthesis by decreasing pruning severity during the dormant season, had little effect on fruit-set in Vignoles. The (UP) treatment decreased yield, due to reduced cluster 129 formation per shoot. The clusters that did develop had fewer flowers each year and thus, less berry set. This was caused by either poor bud maturity the previous year, or because of slower shoot development early in the growing season. Reductions in Botrytis infection were observed only in 1993 in the (UP) treatment. Cane length had no affect on fruit-set orBotrytis infection. Competition induced by fruitful sinks was limited because of the small size of Vi gnoles clusters. However, increasing sink competition among fruitful and vegetative sinks through anthesis was detrimental to vine maturity. Reducing dormant season pruning severity resulted in more vegetation pruned from the vine after anthesis. This resulted in poor crop ripening and wood maturity as less canopy was available for a source of photoassimilate after anthesis. The number of shoots per vine of the (2N), (4N), and (UP) treatments were reduced to the (N) level based on vine size after anthesis. The canopies of differing treatments with the same vine size contained equal numbers of shoots from count positions after summer pruning. Individual shoot growth decreased as the pruning severity was reduced. Thus, with decreasing dormant season pruning severity and increased post-set pruning, the eanopies contained smaller shoots with less leaf and lateral growth by verasion. The (UP) vines could not replace the deficiency in vegetation caused by post-set pruning. Yield, fruit quality, vine size at harvest, and cane cold hardiness, all decreased in the first year. The (4N) vines were affected in the second year showing the same characteristics as the (UP) vines. Cane length had no affect on shoot development or maturity. Longer canes were more productive and fruitful because they were better exposed and produced more numbers of shoots per cluster. Post-set pruning does not appear to be a viable cultural practice for management of Vignoles grapevines in Michigan’s climate. A method to increase earbohydrate competition (through anthesis) to decrease f ruit-set and Botrytr‘s infection and yet, allow for adequate vegetation for fruit and wood maturity for the rest of the growing season, needs to be developed for Vi gnoles grapevines. go Literature Cited . Baker, EL. 1982. Chemistry and morphology of plant epicuticular waxes. In: Cutler, D.F., Alvin K.L., Price, C.E. (Eds): The Plant Cuticle, 139-161. Acedemic Press, Toronto. Bravdo, B., Y. Hepner, C. Loinger, S. Cohen, and H. Tabacman. 1984. Effect of crop level in a high yielding Caragnane vineyard on growth, yield, and wine quality. Am. J. Enol. Vitic. 352247-252. Brown, K., D.I. Jackson, and GE Steans. 1988. Effects of Chlormequat, girdling, and tipping on berry set in Vitis vineifera L. Am. J. Enol. Vitic. 39(1):91-94. Bulit, J., and D. Dubos. 1988. Botrytis bunch rot and blight. In: Pearson, R.C., Goheen, A.C. (Eds): Compendium of grape diseases, 13-15. APS Press, St. Paul. Candolfi-Vasconcelos, M.C. and W. Koblet. 1990. Yield, fruit quality, bud fertility and starch reserves of the wood as a function of leaf removal in Vitis vinifera - Evidence of compensation and stress recovering. Vitis. 29:199-221. Coombe, B.G. 1959. Fruit set and development in seeded grape varieties as affected by defoliation, topping, girdling, and other treatments. Am. J. Errol. Vitic. 10:85- 100. Coombe, B.G. 1962. The effect of removing leaves, flOwers and shoot tips on fruit set of Vitis vinifera L. J. Hort. Sci. 3721-15. Coombe, B.G. 1973. Regulation of set and the development of the grape berry. Acta. Horticulture 34:261-273. Hale, CR, and P.J. Weaver. 1962. The effect of developmental stage on direction of translocation of photosynthate in Vitis vinifera . Hilgardia 33:89-131. 10. Howell, G.S., T.K. Mansfield and J.A. Wolpert. 1987. Influence of training system, pruning severity, and thinning on yield, vine size and fruit quality of Vidal blanc grapevines. Am. J. Enol. Vitic. 38(2):105-112. 11. Howell, 6.8. May 1992. Pruning and training Vignoles grapevines in Michigan. Vintner & Vineyard. Coop. Ext. Serv. Pub. 6(1): 1-6. 12. Hunter, D.M. and J.T.A. Proctor. 1985. Effects of timing of thinning influence on Seyval blanc fruit yields and quality. Presented at 10th Annual meeting of the Ameriean Society for Enology and Viticulture/Eastern section, Mississauga, Ont. (August 1985). 130 I)...» .‘ "F—fi. _ . A. a 13. 14. 15. 16. 17. 18. 19. 20. 131 Looney, NE. 1981. Some growth regulator and cluster thinning effects on berry set and size, beny quality, and annual productivity of ‘De Chaunac’ grapes. Vitis 20:22-35. Marios, J.J. 1987. Botrytis bunch rot of grapes. Oregon Dpt. of Agr.: The Wine Advisory Board Research Report. June (5)21—6. McClellan, W.D. and W.B. Hewitt. 1973. Early Botrytis rot of grapes: Time of infection and latency of Botrytis cinerea Pers. in Vitis vimfera L. Phytopathology. 63:1151-1157. Nelson, KB. 1951. Factors influencing the infection of table grapes by Botrytis cinerea (Pers). Phytopathology. 41:319-326. Percival, D.C., J.A. Sullivan, and K.H. Fisher. 1993. Effect of cluster exposure, beny contact and cultivar on cutical membrane formation and occurrence of bunch rot (Botrytis cinerea Pers.:Fr.) with 3 Vitis vinifera L. cultivars. Vitis 32:87-97. Quinlan, J.D. and R.J. Weaver. 1970. Modification of pattern of the photosynthate movement within and between shoots of Vitis vinifera L. Plant Physiol. 46: 527-530. Reynolds, A.G., R.M. Pool and LR. Mattick. 1986b. Influence of cluster exposure on fruit composition and wine quality of ‘Seyval blanc’ grapes. Vitis 25:85-95. Reynolds, A.G., and DA. Wardle. 1989a. Impact of various canopy management techniques on growth, yield, fruit composition and wine quality of Gewurtztraminer. Am. J. Enol. Vitic. 40:121-129. . Reynolds, A.G. and DA. Wardle. 1989b. Effects of timing and severity of summer hedging on growth, yield, fruit composition, and canopy characteristics of ‘De Chaunac’ 1. Canopy characteristics and growth parameters. Am. J. Enol. Vitic. 40: 109- 120. . Skene, KG.M. 1969. A comparison of the effects of “Cycocel” and tipping on fruit set in Vitis vinifera. Austr. J. Biol. Sci. 22: 1305-131 1. Steel,R.G.D. and J.H. Torrie. 1980. Principles and Procedures of Statistics: A Biometric Approach. McGraw-Hill Book Company, New York, NY. Striegler, R.K., and GS Howell. 1991. The influence of rootstock on the cold hardiness of Seyval grapevines: 1. Primary and secondary effects on growth, canopy development, yield, fruit quality and cold hardiness. Vitis. 30: 1—10. Weaver, R.J. and RM. Pool. 1968. Effect of various levels of cropping on Vitis vinifera grapevines. Am. J. Enol. Vitic. 19:185-193. 132 26. Weaver, R.J. and RM. Pool. 1971b. Chemical thinning of grape clusters (Vitis vinifera L.). Vitis. 10:201-209. 27. Winkler, A.J. 1929. The effect of dormant pruning on the carbohydrate metabolism of the vine. Hilgardia. 4:153-173. 133 ..005 005:. 0.5.3.2 302 0:00:30 050: 0:50.00 55.... :o..0.0n00 0:00.). .0... .5258.» .8 .o ...... 0... .... $0 a 28:00.0 03.2 n. 0000: 0. u0:0.. .0000: m u..o:0 ”N035" :0 05:0.n 0550 .00 50:0. 0:00 N 00.0 to :0 0:.:....n .0 0N.0 05> :0 0000: .2. 0050.0. 00002 .. .NQn (N :0 05:0.n .00 .00n .050 0050.0. 00002 x 0: 0: 0: 0: 0: 0: 0: 0: 0: 0: 0N N 0 0 0 00 N0 0N0 N20 00.0 0:3 00:05:: 0. I N 0 0 00 00 00.. 0.0 00.. :20 025.55 0N NV 0 FN 0N N0 F0 0N. 0N0 30 0:0.— 0. 0.. m .0N .N N0 00 00' .N0 00.? 20:0 20 0. F. N 0N 00 N0 0N 00 00.0 No.0 0:0. 0. 0.. N 0N N0 F0 00 00 00.0 00.0 00:0 ZN 0 0 I 00 ..0 N0 00 00 2.0 00.0 0:3 0 N. I .0 0N N0 00 00 00.0 NN.. 00:0 2 50:3 0:00 x 0050.00. 00002 0: 0: 0: 0.. 0: 0: 0: 0: 0: 0: cw : 0 N 0N N0 0N 0.: 00.0 00.0 0:0.. 0F N.. 0 .N MN 00 00 00. 00.0 00; 00:0 N 50:0.. 0:00 . . 0: . . . . . . 0: 0.. . . . . 0: 0 ON 0.. 0 N 0 0 0 0 ..0 00 0 000 0 2.0 No; 00:05:: 0 0. 0. 0 m n .N 0 0N N0 00 n on. 0 0N.0 00.5 20 n 3 I :N 00N 0 00 ..0 0N 000 0 No.0 00.0 ZN 0 0 o. 0 Z 0 00 00 .0 00 N0 0 N0 0 00.0 v0.0 2 .0.+0.uz. .. 0050.0: 00002 00:00 00:00 00:00 89.0250. 0002150. .00502 x .00..00n 0.555 .0 005300.. .0520... 205.00; .....000 .....000 .....000 05.. 05> 00 050 05> 00 80:0 0050.0... 0050.0... 050 05> 0N.0 05> .oon 0.:0.n000< .:0__00xm 000:0>..0.000> 000:0)..0.000> 00002 00002 N00 P F005 .000. .0. sea... 03:... .0222... ..o__0> Eon. .0 00:_>0n0.0 00.0:05 .0 05.000.00.05 0:00 0:0 .oo:0 0:0 0N.0 05> 0:. :0 50:0. 0:00 0:0 0050.0. 0000: .0 .0050: 0:. .0 00:02.5 .F 0400... ..00... 00:00 0.5.3.2 302 0:00:00 050: 0:50.00 55.! :0..0.0000 0:00.... .2. 202.000 .2 .0 .... 0.. .... $0 .0 200.200 822, n. 0000: o. «0:03 .0000: m u:0:0 ”00.0.0 :0 055:: 05.00 .00 :.0:0. 0:00 0 00.0.0 :0 05:20 .0 0~.0 05> :0 0000: .2. 0050.0. 00002 > .00:0.N :0 05:0... .00 .000. .0..0 0050.0. 00002 x 134 0: 0: 0... 0.. 0: 0: 0: 0... 0... 0: 0. 0. N 0. 00 00 0. 00. 00.0 N..0 23 0222: .. .. 0 N. 00 00 0. .0. 00.0 0.0 :20 0222: 0. .. 0 v. 0. 00 N0 N0. .00 0N.0 20.. 00 0 0 0. 0. 00 00 00. .00 0N0 :20 20 .. 0. N. 0. N. 00 00 .0 .00 00.0 000.. .. 00 0 0. 0. 0.. 00 00 .00 00.0 :20 20 0. 0. 0. N. 00 ... 00 00 .00 0N.0 20.. 0. 00 .. 0. 00 N0 00 00 N00 00.0 :20 2 ...0:0.. 0:00 x 02.0.... 00002 0.. 0... 0... 0... 0... 0: 0... 0: 0c 0... 0. 0. .. 0. .0 00 00 0N 00.0 00.0 0:3 ... 0. 0 ... 00 00 00 0N 00.0 00.0 :20 N 50:03 0:00 .. .. 2 2 . .. .. .. . .. 0 0. 0 .. N N. 4 0 00 0 00 0 0. 0 00. 0 00.0 0 0.0 020.23 0 .0 0 0. 0 0. 0 0. 00 00 a 00 0 00. 0 00.0 0 0N0 2.. 0 .. 00 N. 0. 0. 0 0. 0 .0 0 00 0 00 00 .00 0 N00 20 0 .. 000 0. 0. 0 .0 0.... 0.0 0 .0 000.0 000.0 2 .0.+0.uz. .. 02.200 8002 00:00 00:00 00:00 .00:0..50. 000550. .00502 x .00-.000 35:0 .0 .0523: .0033... 505.00.: 3:030 3:030 3:030 ONE 0:.) mm ON.0 o...) 8 .00:0 0050.00— UOESOC ONE 005 ONE 0....) .OOn. 0.9030000. .co__00xm 00000>=20u0> 000.002.2000) 00002 00002 0 Q 0 w Na 0 _. .000. .0 :20... 0200 00.2.2... 30> 2.0 ... 005.6050 00.0:05 .0 05.000.00.05 0:00 0:0 .00:0 0:0 00.0 05> 0:. :0 :.0:0. 0:00 0:0 0050.0. 0000: .0 .0050: 0:. .0 00:05.5 .0 030...... 135 .60... 093.”. 0.95.2 302 95830 as»: 3528 55.3 5:833 53.). .65 285cm.» .2 .0 .... x; .... «.m a 28:2? 329, ... .33: or "93.. .33: m utozm “$329 :0 3.5:. 3.5.. .3 59.2 280 ~ No.65 co 33 was no woman .2. 3582 $8: .omfioa > m: m: as m: m: m: . m: a... «N4; om omd wad av mdv u B? NF... mmé 98.. 3.53:: n“: 5 rod voé ov 9; no mm mmé mfu :95 09.295 omK 5F mob mm; ov m.mm an no vmd mmé 95.. nub nmv Fwd mm; on fine 3% mo cod vmé tozm zv mws we. no.0 mm; —m 7mm can mm 05m 5... 9.3 mw.v 3.: Sb ...mé mv m.mm noon :N and team ZN 56 X... «r.» mm; mm ...vo non mo and mo... 98.. no...“ 9: 56 mmé on 060 a 5 owd mod :95 z £ch 930 x 3533. 8.32 m: 2 . m: m: m: m: m: m: 2.0 a: 8m cm. 3. m5 3 3a «a.» 9.3 was s: :6 mmé xv mdm ow Fwd no... :95 ~ 596.. 980 .. .. 2. .. mc .. .. .. .. m Swe— n vm 9 Cd 9 No; 3 n odv a me n 5% n cod 09.5.95 3 mm.m a vi a mmé m omé mv a 0.00 a .R a and a mud 2v 9 ate a on; a and 0 mm: mv a m.mm am mm a mvd a mad ZN a own a mm? a an m cm; rm 6 vdo a R a and a mod 2 Amw+mwuzv > 3:33. $32 .38 we; .82.». .36 was .83 ..3 8520 2:9 3 89.5: 32.: 35>.me E253; «3:223 :28 22>. 83.223 >:mm 325m .286 .236 22> 22> 28.. no.0 $2.352”. 33:032.”. .39.? .60 3.8.5: $29.05., >23) :5“. a «655%... 3.965 .0 22> .0 9:38:30 05 23 22> o5 :0 59.2 2.8 new 8582 mono: .o .383: o... .o 8:02.... .m m..m<... 136 .58. 093.... 295.2 :62 9:850 922. 2.528 55.3 5:933 522 ..ms 282:2... .2 a ...... 2.. .... 2.... a 28:22» 322, u. .33: or no.3. ”mane: m ntozw ”ween co 92:3... 92:6 .3 59.2 050 N .025... cc o~.m 05> no 33.. .2. 3532 auto: .8232 > m: .....c m: m: a: w: m: a: an 2. ta 8; 58 one. mm; mv chm I 3.0 vmd 95.. 35:95 to «N. f3 86 0N; No 59. m ..m.o vmd :96 35:95 «A mwé 9mm o...m mvé mm Emu ..m Pm; oNN 9.0.. m4 omh mNm Chm :4 ..m v.2. mN or... moN tozw Z¢ v... we... mNo mad >04 mm mNo vn mm; 5N 9.6.. we :2 mSm we...” 9N2. 0.. 0.: mm on; SN :95 2N ...F no.5 ...S and 00% Pm m.vm v¢ moN and 9.0.. m4 31m va 0N.m vm... mm was cc 2.; no.0 :95 Z £95.. 280 x 853mm 8qu . m: . we we m: m: we m: me N; 3.6 mdm um... mm; mm v.3. on Na; omN 9.0.. ...m Nm.v NKV 55 Na; mm _..mm 5N mré SN 29% N 26.5.. wcmo .. .. .. m: .. .. .. 2. .. .. m mm 9 av; a Q? moN n mNé 9 mm n 9% n or o .56 o and 35:95 a v9 aw mom. a New 3...” n mNé a mm a fmn m on n vNé 5 SN zv n we a 86 a «.8 min a Re a mm a NE a an an mt. on and 2N a me a mud n f: moé a N»... a mm a Vow u Nv a 54 a wad z 639%. > 8:35.". $qu .236 SE was 8 Sosa .26 25> .53 .5. .0620 .E9 ..2. .wnEsc .5... .0539: 2.253.... .285 22> ..EB 22> «8:22: 3.8 32:8 .236 .236 22> 22> 98.. no.0 32:32.2“. 3.28322 .nmmw... .50 US$33... 35.6%... «2955 .o 22> .0 3:32.88 ucm 22> o... co 59.6. 2.8 new 3582 mono: .o 898:: 05 .o 852...: .mEgoc; >o__a> c5... 8 .v m..m<.r 137 ..moh 8.3m 22.5.2 302 9:855 9...... 2.828 5...... 5.6.88 2.8.2 .8. 2.82:2» .8 .o 2... 2.. .... ....m .a 282:2... 822. u. 88.. o. n83 .88.. m ":96 N22... 8 22...... 8...... .8 59.2 2.8 N Nam. .m. 2.... ..o 9.5.... .8 $0.. .23 3529. $202 > 2.0.6. 22.. 05> o... :0 20.8.0 .0 0.08783 2... ......omémuv .3893 ....oogmnm 383". 22.; 9.25.. m. a 2...... 8.2.38 222.83.... 22... .o. .22.... 2._> x .28. 2o: .22.... 2.. 2...... «2...... .0 $8....on ...... ”9.08.83 ”2.8-9.3 .qoovéwum £86“. 22.; 9.25. m. a 2.2. 8.3.3.. 222.833 3... .o. .22.... 25.2. ... .Nm.o..m ..o .52... 223.0 8.0.55 Na .0 2.8.5.38... £95. ...... 2.58 ES. ova... 2.3.38 a :0 8...... 28.5... .032”. uo§==wm > m: w: w: m: m: m: w: w: m: ... .... a ...... o... om... .... ... 8m 2.2 85.95 m. .... .. a... w... t... ... om .mm .55 8:295 .... .... w 98 3.. 8... ... .m .3. 9.3 m. .... a. .28 8.. ...... ..m ow NS .85 2.. .... .... .. 0.8 ...... mo... ..m a. ...... 9.3 .... m... m. ..om ..m. ...m mm mm NS .36 2m a. ma ... 98 on. B... .... m. 2.. 2.3 .... m... m. ..om 8.. co... .m m. man .35 2 59.3 2.8 x 822% 882 we. m: w: w: m: m: w... m: . .... .... .. 0.8 mm. 0.... mm a. mam 9.3 .... mm m. ..om R. ...o om om m... .55 ~ £95.. 950 .. . 2. .. . .. 2. 2. .. a m. n .... o. .. ...... a 3.. a o... .1. a. a .8 8.5.9... o w. a m... .. a com a mu. .... «.... mm om a .8 2.. a M... a m... m. a new a mu. .... 8... mm .m a 8... 2m m .... a ...o ... a mom a mm. a 8.0 3. m. a 2... z .m.+m.uz. > 8222. 882 x “cm 3 “cm 292mm .0520 wb=om .92.; In 5520 2Qan 5w :2... > .8532 2.0539... 22> .2220 ES. .2. 2. 236m 2222.: 32:8 .....25 .32.... .23.”. ......m Na? .o. .80 3.83m... Na. .9 9...... ..o :93. 9.82.22.on Eoo.m $5.636 8.953 .0 >52... new .3 ...... ..o 59.... 9.3 new 3520. mono: .o .35.... m... .o 8.83:... 22.26:... 3...; econ. .a .m m..m<... 138 ......» macaw 0.9.52 :62 9:855 9...... «5.5.8 55.! 5.3.38 2.32 ...... 285...... .9. .o x... .... x. ....m ... 9.8%.... 82.... 3...... o. "9.0.. ”mono: m ":25 ”8.0.9 .... 8.5.... 952. .8 59.... 250 .08. ..m. 2.... ..o 95...... .8 88 .2... “.0522 882 ...2.o. 2o... o...> m... ..o flaws... ... ....oowéfim ...... ”$8.8". _$oo.o..nm .....o..-o~nu ”$83. 29.... 9......2 .... a 9...... 8.3.9.. 2389...... 8; .o. .22.... 2.5 62.... 2o... .83.". o... 5...... 3.8.. .0 ....oopéwum ...... gowéouv ”9.8-3.4. .....ovémuw ”$8-0". 96...... 9.2:... m; a 9...... 3.3.96 202.83.... mm... .... 5.8.0 .......>> .... NH. > X @285 ..o 5...... 229.6 $.95.) on .0 35.5582 595. ...... 2.58 ES. cum... 92.5.» a co com... 235.... 526E no.wE_.wm > w: m: w: a: m: a: w: a o..~ a... «No 3. i new 9.2 85.95 .. ...om .... mm... am I own ......m 8.5.9.: m. ...om m... cm... .... ... ...m 9.3 m. ...om .... mm... mm om «on ......m 2.. u. ....N mm. C... S .m ....m 9.3 ... ...Nm 8.. ...... ... 8 «3 ......w 2m t «.mm 3.. ...... mm mm ....m 9.3 mm ...Nm ...... m... 3 ... com .85 2 59.3 2.8 x 852.... 882 m: w: m: w: m: m: w... ... «.3 a... ...... om om NR 9.3 i ....m ow. ...... ... ... EN :96 ~ 59.... 980 a o. .... 3m .. o... a 8.0 a mm .. ... .. owm 8:295 .... m. .. ...om .. o... a a... a .m .... ... a mam 2.. .... t a ...m a ..m. .. ...... a 8 a mm a 8m z~ m cm ... ...mm a mm. .. ...... a .... a .m a Sm z .m.+m.uz. .. 852.... 882 2953 5.3.0 «2.8 3.20.. In .250 macaw .90. .....n. > .3632 E2“. .om .... masow 2222.... $2.5m 2.620 28.3. .632“. ..mm 2.2...me dam. .. .00 8.823. .8... .8 0.5.. ..o :93. $5623on 585 $28.23 >o__a> :5”. .a 8:323... 8.0.55 .0 2:2... new .3 .....n. ..o 59.2 9.8 ....u 35...... 3...... .o .35.... o... .o 8:02.... .0 mame 3m 139 mmmdmw + co www.mw .... 09 3:6 u 9%. 3.3 1.52%.. mwkmao NF or m o v m o _ _ _ . F . _ b r bf _ p . o woo_ woom Mu , H 1000 N w n r W , a woov U moon ,oom .mmcSmaSm 8.89) E wEmEmSmmmE .236 8a «526: .2 Ewncmfi «mm. .P .9". I40 :m mat + 3 Rmem u as 88.0 u «E 3.8 Ikozwq mwhmDJO 9 m? 3 N? 0.. m w v N o b . _ — Ll _ ’ — _ _ _ b p _ F b _ h _ b . — _ h P — _ P w o HBSWHN HSMO'H I O O V Y 1' I ‘f'ffi O 8 com .mmc_>mam.m mm_ocm_> :_ 3589888 .936 8a 2950: .2 2%:me mam. .m .9“. I40 , aim Rafi + S Rafi u es 886 u NE @3533 $530 2 .8 3 NF 2 m m v N o . L . h _ p p _ _ p. h _ p . . — _ h p — b h h o l O O N T I 1 l O O CO I l I I HHBWHN HEMO'H § I .mmcswafim 3.955 E mucmEmSmmmE 8620 8a 9950: .2 2355 89 .N .9“. 14] 2mm macaw gem—2 Egan... .mnEmomo - o it. is .3. W o r mm. H H w. n o , m «.02 2 m D. a , I. m m._mm£> >m=m> Emu E $5536 $633 5 A88 £95. 505 meE :0 btm>mm @555 6 85.35 .m 9.... HmmDm 03m EMF”? wxmmg mm ON 9 0.. m o _ _ k . . _ — V L p _ _ — .— — _ _ _ .r P o .— r _ _ low .19. m ZVI 2| . mm .533 @5265 mm? 9: mczau m9m>¢£> >m=m> Emu. E $5535 $635 5 cozoznoa $2 80% meE :0 3:93 9:55 ho 852E. .m .9". J.GOHS Had SSAV31 145 mm ...mmDm 03m mmt< mxmw>> ON 9 o_. m o _ p T _ _ _ _ . _ . p . _ . _ _ _ _ _ , o wm WV .10 0.0 - 0.0 - -_ .. illiiik ...m a wow 9 n5 II ZN I - m m f 3 .:ommmm 0:566 83 9: acts: 3.96:3 >m=m> ::m.... E mm:_>mam:m mm_o:m_> :_ :ozosuoa Socw .993 :o 3:98 9:5 .0 853:. N .9". HEQWHN 'IVHElV'I ...mmbm 03m mwt< mxmwg 146 om m. o. m o L F p b . _ _ : _ _ _ . F _ _ _ r . _ _ _ _ + .o £11 #11 film I: 0.0 m we 0 -\\-\\\\ ,.. . \ wow 0 9 Km .8 n m .8 ”.8 a — “5 fl 2v II. 2m I 2 II .9 ,8 .:ommmw @5266 39 9: 9:3: «2:583 >m=m> :5”. am 35585 mm_o:m_> :_ :ozozcoa Km. .92.“... :o 3:93 @555 :o 852E. .m .9”. HBGWHN :IVE'] WVHELV'I 147 mm HmmDm 03m IMF”? mxwm>> ON 9 or m o _ _ H _ h . _ h b — p _ p L h p p h _ _ . _ p o «i \ - m fanizvllfillzI; our .:owmmm @5265 mm? 9: acts: «Eggs >m=m> ::mu_ um mm:_>wam:m mm_o:m_> :_ AEov 50:2 82w .895 :o 5.9% m:_::a .o 852E. .m .9“. (UJO) H.LEDNE'I .LOOHS "IVHBLV'I ...mmDm 03m mmE< wxmmg 148 mm om m: o. m o p _ _ _ _ p _ _ _ _ + _ r _ b _ h _ _ _ . _ L p O - i -8 . m . V 3 .I r m :29 wow. u o - M: o H w wooman n . H , .o. - wommrw. m mall2¢||z~|zll_w 8m .533 @5265 mam. 9: acts: «298:3 >m=m> Emu um 3:30.36 «2035 .o 9999+me28 80% :8 :65. m2§mmm> .59 m5 :0 2:98 @555 .o 852:. .9 .9“. 149 mm meDm 03m EMF"? wxmm>> ON 9. 0% m _ — h p p k _ b r b h b h b — b p r p O «a. lellZleNIIIZI lIIIIflITjTTIIITYITjIT'IIfYIITrTITTIT .:ommmm 9.39m 89 9.. acts: «963:3 >m_.m> :5... am 35586 8 8 8 8 2 ° .I.OOHS Had saAva'I 1V.LO.L O (D O [\ wm_o:m_> .o afimfiibméav 60% :mn. 82$. .99 m5 :0 5:98 9.55. .0 853.2. .2 .9... CONCLUSIONS 150 151 The data presented here leads to the conclusion that the competition among fruitful sinks (clusters) during anthesis has a greater influence on reducing f ruit-set than vegetative competition. Lower berry set provides a loose cluster architecture and thus reduced Botrytis bunch rot. Reduced pruning severity and post-set thinning produces grcater numbers of fruitful sinks during anthesis, resulting in a greater response in Seyval grapevines. Reductions in yield due to decreased fruit-set can be overcome by increasing the number of clusters thinned. Thus, the level of sink competition can be manipulated by pruning severity without adversely affecting yield. However, the vine’s capacity limits the level of vegetative sink (apical meristems) competition it can accommodate and mature both fruit and wood. This is also evident in the Vignoles cultivar. However, because they bear smaller clusters, the fruit-set response to greater fruitful sink competition is less notable. In order to reduce berry-set sufficiently in Vignoles and improve cluster architecture, pruning severity must be reduced to the point where vegetative sink competition adversely affects vine health. The hypothesis discussed in the introduction concerning reductions in fruit-set to reduce Botrytis infection may have some practical applications as to how large clustered cultivars are managed, but appear to be impractical for smaller clustered cultivars. Hedging Seyval grapevines increases the number of nodes within the canopy cach year. Greater numbers of shoots increase early filling of the trellis area with foliage, but little leaf area production occurs after veraison. Thus, older leaves are responsible for photosynthate production during fruit and wood ripening. Due to the early leaf arca production, hedged canopies become shaded by the time of veraison. The shoots are considerably smaller and grow upright for much of the season. This contributes to leaf layer formation which increases canopy density and shading, reducing cane cold hardiness. Canopy density also promotes Botrytis infection. If hedging is the preferred pruning method, then cordon renewal is recommended after four to five years. 152 A moderate pruning severity in Seyval grapevines (30r40 in Michigan) that adequately spaces shoots along the cordon allows for early filling of the trellis area with foliage. Shoots are more developed before bloom facilitating berry development, resulting in greater yields. Shoots produce laterals after bloom that bear new leaves during fruit ripening and wood maturity. The optimal shoot spacing inhibits leaf layer formation and shading is rarely a problem. Thus, cold hardiness is improved and Botrytis infection reduced. Optimal shoot spacing within the canopy should be the primary consideration during pruning. If mechanization is necessary to reduce labor costs, then mechanical fruit thinning needs to be developed for highly fruitful cultivars such as Seyval and Vignoles. Hand pruning to ensure proper shoot spacing and sink competition, along with mechanical fruit thinning, would be a more cost effective method for controlling Botrytis infection and maintaining vine health of these cultivars. ”(J a“! ‘1.) ‘ APPENDIX I INFLUENCE OF CANE CHARACTERISTICS ON PRIMARY BUD AND PERIDERM COLD HARDINESS 153 Summary Striegler and Howell (1991) demonstrated a clear response in cold hardiness to varying cane characteristics in Seyval grapevines. Three classifications (excellent, acceptable, and poor) were developed from this work and from hardiness evaluations made in December of 1991 and January of 1992 on canes from Seyval grapevines. Primary bud and periderm TSO’s were determined for variations within four cane characteristics, and over node positions along the cane. Canes having a medium diameter (7-10 mm) (Figs. 1 and 2), medium intemode length (6-8 cm) (Figs. 3 and 4), and dark periderm (Figs. 5 and 6) were found to have superior primary bud and cane cold hardiness. Persistent laterals (Figs. 7 and 8) and node position (Figs. 9 and 10) did not have any influence on cold hardiness. Samples of canes within the three classifications were evaluated for cold hardiness in December, January, and March following the 1992 growing season. Excellent canes were superior and poor canes were inferior to acceptable canes in both primary bud and periderm cold hardiness at each date and over dates (Figs. 11 and 12). These results were similar to those found by Striegler and Howell (1991). Using the three classifications was a rapid and effective way to evaluate the relative cold hardiness of each treatment vine (see Chapters 1 and 3). 154 155 .m um E 30 m : m 0 o m ...... 1 0000000000000 oooooooo fl ........... .......... . .... r ooooooooooo .......... ooooooooooo .......... ----------- .......... ........... .......... ........... ..................... “v n .................... _ .......... ........... .......... ........... .......... ........... P ........... ..................... .......... ........... — .......... ........... ........... pm ........... E — t ........... .......... ...... D .................... b nnnnnnnnnnnnnnnnnnnn uuuuuuuuuuuuuuuuuuuu .................... m oooooooooooooooooooo Wk ' O . MI— OOOOOOOOOOOOOOOOOO o-o-u ooouuuoooconnuuco-no VUIHUHOHIHIHOHIHIHIHL —. l 5.. n .................... N u m m m e. 2 * ooooooooooo i OOOOOOOOOO - .......... v .......... r 1 .......... z a... f ................... 4 .......... j ........ nnnnnnn 00000000000 ....UIH . O ..... ......... ............. OOOOOOOOOOOOOOOOO ' s .... .. _ .w.. ......I......'.....- W 00000 ....................... T .. ...................... Y .................... Y 00000000000000000000 I I 000000000000000000 .............. .u 0000000000 I ...... .- fl I I I 1 PO - m l r _. a - .m 1| F 1 m l O . r r 1 VI T T T .m; m w .o m mob. : E .U. o m .. FA U Nm .0 m o . 3.3 a ”E a E m E o tQ—i 0““ “5E3. OWN—”...... m E .U n o v : won—f no E mom : .m o o: a ...: $8 . .m .n. (o ‘a m 2 30'68 S5. 1690):, J. On 0 8'- M B 0.: Ltd 8 8 EoEmE mcwo . Ems. . _ .i DOE _ .D (II C .2. Nam...NN Z<... 59. .9 Own. 8 fiWjI'll mv .88 9A $93 .EE 94. ”E232 .EE nv ”..mEm $83235 .mimm .o $3.2m: 28 552.3 9.. :o .2955 28 ..o 8:022. .m .9". (0192 MOIGQ 3991690) 091 wrapped 157 590.. 80:52. 95.. _ 8.2 _ :95 r t t .85. _ :Bm ........... o ...H...H..HH..HH...U. .... f Whammy”... ...”...mwun ”...”...mmm. . ”....fim.” .UH.HHH.HHH.HHH.H..H madman“... H mime... WEE”. ”“wame ..m .....H..H..H... ...”...mww H.....HH..HH..HH..HH . d ....................................................... r U. H..H..H..H..H.. .......HH..HH..HH..HH H....H..HH..HH..H.. w “mum“... .mwmmmm WWW.» ..op m H....HHH.HH.HH._H ”.3.me ...“.Hunmmnvu. . 8 ............................................. n ”......HH..H....H.. ...”...HH..HH..HH..HH HH...H..UH...H..H.. . D. “.mmmnmwnww Hummumflmm.” ..H..HH..HU..H..H . I. .....H..H..H..H. .....u..u..w.nu. III... fimw W unnamwmm HH......HH.HHH.HHH.HH m ...”...wam ”...”...H... nu ..HHMH n . ”...”.wmw.” Human.“ . PM ...”...Hmm.” m..u......u..w.u ION % .H..HH..HH..H..HH. ..HH..HH..HH..HH..HH. . s ”.....me ”Hummuwww. . mm “unsuswhn. ..HH..HH..HH..HH..HH. T m ...”..H..H..... -mm 2 . m . o ( am a a . now ¢ . v Nmmw .mm 25. 59 .2 Own. . . T mm .20 R ”9:3 .20 mé ”22.8.2 .20 av ”:05 883285 22.9w 2 8862: 28 :3 2952 on. :o :58. 80:82. 2 8:8.2. .0 .9”. 50:0.. 80:52. 158 9.3 _ 8.). _ :20 _ t 08.. _ .85. r :20 0 En...” ......”un .....ww... ”Hyena”... 1.. .m.m.m.m.m. nymmwmmw.” ..HH.H.HH.N.HH .HHH.HHH.HH..HH.HH.. H wuumumuflm mmmmmmmumu. Wmuhwnnmu. ”.HH..HH..HH..HH..HHU . d ............................................................................. . - .......H..H..H..H H...H..HH..HH..HH..HH .....H...H...H...H..H ”...“...u..u..u... .I or m. ...”...Hfi. ...”..undm...“ ...”..H..H..H..HH .....H..HH..H..H. . P. . H...H...H..HH...H.... H a ......”wa ”Humnunwfi mummy. mum»... ”.9 w r .... . I. .............................................................. - 9 ...... ..... . .... . 0 .... . ................... l \I wwwnumwn .HH.HHH.HHH.HHH.HHH.H ”hwuwwnwfi WHHHHHL . ON a mfi.~.~.m Juana”... .....H..~..H..H.. ...”..unmuwmm. .m ............................... . .............................................................. . . 8 .Hn.m.m.m._uu. .HH..HH.HHH..HH.HHH.. ...”.mmmmw. n. N % H..H..H..H..H..H H.HHH.HH..HHH.HH..HHH ....lllIL . 8 ”an...” ”Rum... .. P ..... rill .1 on m , m w mm m. .. . 5.9. mmmp .NN 23.. amp .2. Own. H r mt .20 mA ”0:04 .20 0-0 ”2200.2 .20 ov ”:05 083805 31080 88.0.0: 0.00 2:00:00 05 :0 50:0. 80502. ..0 8:82. .v .0.". 159 5.00 2.02.9. :305 0:00 _ :306 Em... _ _ :305 0:00 _ £505 Em... ............................ . . III: 1 .................... A oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo ...H.”.......m.w.uwu ...HH....H.H...HHH.HHH.HHH.. .H...”......H..H..H..H.. ..HH......................H. . mm“wmmuunwwm Hmmmummmmmum” nnHHHHHHNHHHHNHU HumHummuuuumuun .l . ...an...”......H......, ......H..........H...”...HH. .”......H......H..H..H..U u..........H...........H.. . ............................ . ............................ v.........o...o.ooo...o...... .........o....oo.o..oo.a.... I ...H...H...H...”swamp”. ........................................ . ............................ ..................... . . .................................................................................................................. 1 mm...”.....H...H..H..H H......HH..H............H... ......H..H..”......Hm. ......”...”...u......u...U . ...“......wmmmmH” ”unuwwm”mi... .H”...”...”mummy.“ H.m.”WNW“... . ...HH.....H............ we.........0.” ..H”......“..H..H..H..H. ......HH...H...H.....H...H . an...“......H..H.. ......H......HH.fix”. .............H..H..H..u..u l.-- ..... .. .... . . ..................... A ... u..- -.- .o. ... ... as. 1. on. no. you can an. on. n. T ...........H..”xv...” H......H.......HH.......H... ..HH......H..~..H..H..u. . .. ..............._...p.. ..... . r Wham.“”human” H”.mpummmuxman.” l ...H.........H...H...n.m ”...”...mm.........HHH. . a...”......HHHHW. l- . “......“”...”?me . . . W.M.”..wm.mums” a .................... . ... .. ..u' . .. «mm. .mm 23. 59 9 Own . . 8:32.05 .m>>0m .0 88.0.0: 0.00 000 20:20 05 :0 .0.00 2.00:8 .0 8:82. .m .0.“. a a :2 2 (ciaz moles $891560) 09; png Mewud O (O In (0 160 :30.m 0:00 :30.m Em... . _ 5.00 2.02.01 .............. .............. .............. .............. .............. ............... .............. .............. .............. ............... .............. .............. ....... nnnnnnnnnnnnnn uuuuuuuuuuuuuu .............. .............. ............... .............. .............. .............. ............... .............. .............. .............. ............... .............. .............. oooooooooooooo .............. .............. .............. .............. .............. .............. .............. ....... ooooooo 00000000000000 ....... ....... oooooooooooooo ooooooooooooooo oooooooooooooo .............. uuuuuuuuuuuuuu uuuuuuuuuuuuuuu oooooooooooooo .............. .............. uuuuuuuuuuuuuuu .............. .............. oooooooooooooo 000000000000000 nnnnnnnnnnnnnn uuuuuuuuuuuuuu .............. .............. uuuuuuuuuuuuuuu .............. .............. oooooooooooooo oooooooooooooo ooooooooooooooo 00000000000000 ....... 00000000000000 IIIIIII oooooooooooooo ------- 0000000 oooooooooooooo ccccccc .............. .............. --------------- oooooooooooooo oooooooooooooo .............. ooooooooooooooo cccccccccccccc oooooooooooooo 00000000000000 ............... oooooooooooooo 00000000000000 nnnnnnnnnnnnnn ooooooooooooooo oooooooooooooo .............. oooooooooooooo ooooooooooooooo oooooooooooooo cccccccccccccc -------------- 000000000000000 .............. .............. ....... ....... oooooooooooooo oooooooooooooo 00000000000000 ....... ....... ........ ....... ....... ....... ooooooo nnnnnnn ooooooo ccccccc ........ ....... ....... uuuuuuuu ....... ooooooo nnnnnnn ooooooo ....... ....... «i Nmmw .NN Z<_. £505 200 _ £505 Em... . . ........................... . ........................... L . .................... . ..................... w 000000000000000000000000000 A 0000-00--coco-conconoQ-Hoocug ........................... . 1 ........................... . ........................... L ............................ . ...........................L ............................ . L 000000000000000000000 ‘ .0. on. can 000 000 no. 00- ......................................................... v .......................................... . ...... .. ..................... . zxaaaxx ............................ . . ........................................................ . ........................... . ............................ oooooooooooooooooooo ‘ no. on. 000 In. no. no. on. ............................ . . . ........................... . ............................ ............................ . ......................................................... ...H...H...H...H...”...”...HJ ..”...“..ouo..Hc.-H...H.-.Ho IIIIIIIIIIIIII A .0 O. I. II .0 I. O. ............................ . ............................ . ............................ . ............................. III w: 52. .2 Own. T] I T III'TIII'W—fir IO 0 F P (\I (0va mouse 8661590) 091 weaned 7 I O (O .--...” 8 8 I I .0>>0m .0 88.0.0: 0.00 2.02.00 0:. :0 8.00 2.02.00 .0 8:022. .0 mt 0830005 ID v I6] 0306 2.0.0.. 20.0.0.9". OOOOOOOOOOOOOO 00000000000000 00000000000000 ------- ....... 00000000000000 IIIIIIIIIIIIII OOOOOOOOOOOOOOO oooooooooooooo .............. 00000000000000 ooooooooooooooo oooooooooooooo 00000000000000 oooooooooooooo IIIIIIIIIIIIIII 00000000000000 oooooooooooooo .............. OOOOOOOOOOOOOOO 00000000000000 00000000000000 0000000 ....... -------------- 00000000000000 oooooooooooooo oooooooooooooo uuuuuuuuuuuuuu .............. oooooooooooooo oooooooooooooo .............. 00000000000000 --------------- -------------- oooooooooooooo oooooooooooooo --------------- oooooooooooooo 00000000000000 nnnnnnnnnnnnnn ............... uuuuuuuuuuuuuu oooooooooooooo 00000000000000 ooooooooooooooo .............. uuuuuuuuuuuuuu nnnnnnnnnnnnnn ....... oooooooooooooo oooooooooooooo 000000000000000 00000000000000 oooooooooooooo eeeeeeeeeeeeee ooooooooooooooo oooooooooooooo OOOOOOOOOOOOOO ....... 00000000000000 0000000 .............. ....... .............. 00000000000000 ooooooooooooooo oooooooooooooo oooooooooooooo -------------- ooooooooooooooo oooooooooooooo cccccccccccccc .............. IIIIIIIIIIIIIII .............. .............. ...... oooooooooooooo oooooooooooooo IIIIIIIIIIIIIII 00000000000000 tttttttttttttt oooooooooooooo ....... cccccccccccccc .............. ooooooooooooooo oooooooooooooo .............. oooooooooooooo .............. .............. .............. .............. .............. .............. 0: «mm. .NN Z<0 .............. OOOOOOOOOOOOOOO .............. 00000000000000 ....... ....... 00000000000000 00000000000000 IIIIIIIIIIIIII 00000000000000 OOOOOOOOOOOOOO OOOOOOOOOOOOOO OOOOOOOOOOOOOO 00000000000000 00000000000000 ....... .............. 000000000000000 00000000000000 00000000000000 ccccccc ....... IIIIIIIIIIIIII oooooooooooooo OOOOOOOOOOOOOOO OOOOOOOOOOOOOO 00000000000000 ....... IIIIIII 00000000000000 00000000000000 ooooooooooooooo ssssssssssssss .............. 0000000 llllllllllllll ....... ooooooo OOOOOOOOOOOOOO 00000000000000 OOOOOOOOOOOOOOO 00000000000000 oooooooooooooo OOOOOOOOOOOOOO 000000000000000 CCCCCCCCCCCCCC IIIIIIIIIIIIII ....... lllllll OOOOOOOOOOOOOO 00000000000000 IIIIIIIIIIIIII .............. OOOOOOOOOOOOO ....... 0000000 ’l’ 8 D 8 8 m: Pmmw .9 Own. I ' I I IO 0 ID 1- 1- l O N (0332 M038 8333930) 091 one AHVWIHd I In N I O (0 .0>>0w .0 88.0.0: 0.8 0:: 20:20 0:. :0 8.0.0 .0.0.0. .0 00:022. N .m.u. 8830005 ID ('0 I62 00.0.0 00.0.0. 20.0.0.00 00> _ 02 _ _ 00> _ 02 -i-lll-“ --. v ..... H ...“...H..M.|..H ..... . ............................ 1 o ............................ . . . $TH$$E .Hfiflfifim Vfififl$fi Wfififimw . mmwmmmmmm”mm ...mwmmfim .Hmmmwmmwmm Hm.wH00... .0 ......“ananH... H”a...”W.M.”... ”...”...”Hmnuww. Hm.””W.M.“...mm H d fl...........flH. ”......”...»... ......mmm”My”... H.H.”.....ummm... ”.0. w. ......................................................... . o. afiflfififl. mfifififiw $fifl¢$y KfifiEfim . m . “may......H..H..... ..........H......u.w.... .... . H. mp “awn..HU..HH..H..HH..H. Erna“.”Havana” banana”.Haw.””w H.HH.HH..HH.HH..H.UHH.Hu r W. n....””HE...H. ”......“”......“ ....wa”Hum”... Hmm.in... 0 I R.E....“.... ”in.“Human n”HM...””mm...“H. ....mmmmn”.... H m ............................ .. ... ... ... ... ... ... .. ... ... ... ... ... ... ... “Human...”....H. an...an”... --- .......0.........“ .1 mm % uwmmfiwwm“ mm...””mum... ....... ,. ..... .. H m m.«H.200 H.H...WEE ,. 00 m ”my...“my... H.m.......fiw n a - .. . . . .. mm m . ( 0: 0: .. ow 000. .00 23. .00. .0. 08 . . 00 0830005 .0200 .0 88.0.0: 0.00 2.00000 0:. :0 020.0 .220. .0 8:022. .0 m... :00.00 20.". .00.0< 80.000 000 163 ...... I I I ...... ...... I I I oooooo I I nnnnnn I I I 000000 I I 000000 I I I I I I ...... I I 0 ...... I I I 000000 I I ...... I I I oooooo I I 000000 I I I nnnnnn I I nnnnnn I I I ...... I I nnnnnn I I I ...... I I ...... cccccc I I oooooo I I I nnnnnn I I oooooo I I I ...... I I nnnnnn I I I I I I oooooo I I oooooo I I . I 000000 I I ...... I I I ...... I I tttttt I I I ...... I I ...... I I I I I I ...... V I I ...... I I I ...... I oooooo I I ...... I I I uuuuuu I . ------ I I I aaaaaa I I I IIIIII I I ...... I I I nnnnnn I I I I I I oooooo I I I ------ I I IIIIII I I I nnnnnn I I I IIIIII I I IIIIII I I I ...... I I IIIIII I oooooo I I I IIIIII I I I ...... I I I IIIIII I I oooooo I I I nnnnnn I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIII I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I I IIIII I I I IIIIII IIIIII IIIIII I IIIIII I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIII I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I I I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I IIIIII I I I I I I IIIIII I I I I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I 9 IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII IIIIII I I I IIIIII I I IIIIII I I IIIIII I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I IIIIII IIIIII I a I IIIIII I IIIIII I I I IIIIII I I O IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII o I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I o IIIIII I o IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I I I IIIIII I I IIIIII I I I IIIIII I I O IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII I I I I IIIIII I I IIIIII I I. IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII Q I I IIIIII I IIIIII I IIIIII IIIIII IIIIII IIIIII I IIIIII IIIIII IIIIII IIIIII O o I IIIII O I I IIIIII I I I IIIIII I I I IIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I O I I I IIIII I o I IIIIII I o I IIIIII I I I IIIIII I o I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I I IIIII I o I IIIIII I o I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII o I I IIIIII I I I IIIIII O I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII o I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I IIIIII I I IIIIII I IIIIII I IIIII I I IIIIII I I IIIIII I I I IIIIII I I I IIIIII 0-0 _ .0 I I O IIIIII o IIIIII I O I IIIIII I I I I a o IIIIII I O I IIIIII o I o I I I ...... a I O IIIIII I O 0 IIIIII O c IIIIII I o o IIIIII I I IIIIII I I o IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I a I IIIIII I I IIIIII o I I IIIIII I I IIIIII I I I IIIIII I I IIIIII I I I IIIIII O I IIIIII I I IIIIII I IIIIIII IIIIII IIIIII IIIIII I C ,— I ID F l . IO N ' I ‘ I 8 8 (0192 Moles 8861590) 091 pna Mewud f In G) I m: Nmmw .NN 25. .59. .2 Own. 0830005 .0200 .0 008.0.0: 0.00 0:: 202:0 0:. :0 0:00 0:. 80.0 82000 000: .0 8:022. .0 mt 000.00 .00.“. .029. 002000 03 164 IIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIII IIIIII I I I I I IIIIII IIIIII IIIIIIIIIIII IIIIII IIIIIIIIIIII IIIIII IIIIII IIIIIIIII IIIIIIIIIIIIIIIIII IIIIII IIIIII I I I IIIIII I I I IIIIII I I I IIIIII I I I IIIIII IIIII IIIIIIIIIIIIIIIIII IIIIIIIII IIIIIIIIIIII IIIIII IIIIII I I I IIIIII IIIIIIIIIIII IIIIII IIIIIIIIIIIIIIIIII IIIIIIIIIIII IIIIIIIIIIIIIIIIII IIIIII IIIIII O I O IIIIII I I I IIIII IIIIIIIIIIII I I I I I I IIIIII IIIII IIIIIIIIIIII IIIIII IIIIIIIIIIII I I I IIIIII IIIIII IIIIIIIIIIII IIIIII IIIIII O O O IIIIII I I I IIIIIIII IIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIII IIIIII IIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII IIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIII IIIIII IIIIII I I I I I I IIIIII I I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIII IIIIII I I I IIIIII I I I IIIIII IIIIII IIIIIIIIIIII I I I IIIIII I I IIIIII I I I IIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIII IIIIII I I I O IIIIIII I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIII IIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII IIIIIIIIIIII I I IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII IIIIIIIIIIIII IIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII IIIIII I I I I IIIIII I I IIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIII ’ IIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIII IIIIIIIIIIIIIIIIII IIIIIIII IIIIIIIIIIIIIIIIII I IIIII I I I I IIIIIIIIIIII I I IIIII IIIIIIIIIIII IIIIIII IIIIIIIIIIIIIIIIII I I I I IIIIII I I I I I IIIIII I I I IIIIII I I .i‘ C. Nmop .NN 25.. 5m? .0. Omo 00030005 .0>>00 .0 00050.00 0.00 0000000 00. 00 0000 00. 0020 002000 0000 .0 00002.5 .0. .90 P I In P . I v . no N (0192 MOPS 3991590) 09.L ”113de O CO ATV ID 1') l O V 165 news. — 28 anmm banana _ .3888 _ «g. ,3 «50¢ m.._m<._.n_m00< n hzw44m0xw fifll- 10 I O ,- Ytl1* In F v Y i ' T 1 IO N g g (0 (0192 MOISQ 9991690) 091 ssaugpmH uuapuad I ‘ l0 (0 .mEgmcs >m=m> Emu. 8 $55qu 326w .0 $8.22 28 589.8 :0 moumtmuoemzo menu B 85.55 .3 .9“. 9 166 mama macaw .mEgmcS >m=m> cam“. um «9.3255 $13 .0 «8:65... 38 95 basin :0 85:93.20 memo .o 8522. .3 .9“. 2822 b352, 598800 _ _ L _ _ C t .5 cc :10 ..m o moon. m. n a m A: H , m m .1 m. 52.58% ..mpm a m . 0 o a [cum Fzmjmoxm , .m n m. [mm a n ,. Q m m won . w. W mm HICH IGRN 3‘12 5m: UNIV. LIBRnRIEs IIHIIMWIWIIVWWW"WI 93014150837