Effects of Prohexdione-C on Yield nd Composition 73 Effects of Prohexdione-clcium on Grpe Yield Components nd Fruit nd Wine Composition Dnielle Lo Giudice, 1 Tony K. Wolf, 2 * nd Bruce W. Zoecklein 3 Abstrct: Prohexdione-clcium (prohexdione-c) ws pplied to field-grown Cbernet frnc, Cbernet Suvignon, Chrdonny, nd Seyvl to evlute rtes nd timing effects on fruit yield components nd on fruit nd wine composition. Berries per cluster, berry weight, cluster weight, nd clusters per shoot in the subsequent seson were ll decresed by multiple, prebloom plus postbloom, pplictions to Cbernet Suvignon nd Cbernet frnc. Similr reductions in current seson components of yield were observed with Seyvl. Appliction (250 mg/l) to single clusters of Cbernet Suvignon nd Chrdonny t bloom, or in the one-to-two-week prebloom period decresed fruit set, wheres pplictions one to two weeks postbloom reduced berry weight, with no impct on fruit set. Berry weight reduction correlted to incresed color intensity (420 nm + 520 nm), totl nthocynins, totl phenols, nd phenol-free glycosyl-glucose (PFGG) in Cbernet Suvignon. In seprte experiment, prohexdione-c incresed Cbernet frnc must color intensity, totl nthocynins, nd totl phenols, despite hving miniml effects on berry weight or crop yield. Arom nd flvor tringle difference tests did not distinguish tretment differences with young Cbernet frnc wines. This study of prohexdione-c effects on grpe reproductive development illustrted tht berry set nd berry weight were responsive to ppliction timing, with the one-to-two-week period fter bloom most sensitive to reductions in berry weight. The concurrent effects on fruit composition were generlly positive, while the full impct on wine qulity remins equivocl, but worthy of further evlution. Key words: Apogee, plnt growth regultor, prohexdione-clcium 1 Grdute Reserch Assistnt, Deprtment of Horticulture, nd 3 Professor, Deprtment of Food Science nd Technology, Virgini Polytechnic Institute nd Stte University, Blcksburg, VA 24061; 2 Professor, Agriculturl Reserch nd Extension Center, Virgini Polytechnic Institute nd Stte University, 595 Lurel Grove Rod, Winchester, VA 22602. *Corresponding uthor [Emil: vitis@vt.edu] Mnuscript submitted April 2003; revised September 2003 Copyright 2004 by the Americn Society for Enology nd Viticulture. All rights reserved. Grpe flvor nd rom components re not uniformly distributed in the berry but rther re concentrted or comprtmentlized in specific regions. Glucose nd fructose re concentrted in the flesh, wheres inorgnic ions, phenolics, nd trtrte re concentrted in the skin (Coombe 1992, Coombe nd McCrthy 2000). Phenols imprt color nd structure to wine, while nonphenolic compounds, such s monoterpenes nd norisoprenoids, re lrgely responsible for rom nd flvor (Willims et l. 1992). Flvonoid phenols, prticulrly nthocynins, re found in greter mounts in red cultivrs thn in white cultivrs, primrily in the skins (Fernndez de Simon et l. 1993). Anthocynins re found in the vcuoles of cells locted few cell lyers below the epidermis (Roubelkis-Angelkis nd Kliewer 1986). Nonflvonoid phenolic compounds such s hydroxycinnmtes nd nonphenolic monoterpenes such s gerniol nd nerol re lso more concentrted in the skin thn in the flesh of the berry (Fernndez de Simon et l. 1993, Prk nd Noble 1993). A decrese in berry size would led to n increse in the surfce to volume rtio, theoreticlly incresing the proportion of these flvor nd rom precursors in the must nd wine nd potentilly incresing wine qulity. Smller berries cn be clonlly selected nd cn be chieved by deficit irrigtion (Coombe nd McCrthy 2000, Estebn et l. 1999, Reynolds nd Nylor 1994) in rid regions. Aside from potentil increses in wine qulity, reduction of berry size nd cluster compctness my lessen the incidence nd severity of botrytis nd other bunch rots, prticulrly in humid environments. Vil nd Mrois (1991) found significnt correltion between cluster compctness nd the susceptibility to bunch rots cused by Botrytis cinere. Prohexdione-clcium (prohexdione-c) (3-oxido-4- propionyl-5-oxo-3-cyclohexene-crboxylte) is gibberellin biosynthesis inhibitor currently mrketed by BASF s Apogee. Prohexdione-C is currently EPA-registered for pples (Byers nd Yoder 1999, Owens nd Stover 1999) to suppress vegettive growth. The mode of ction of prohexdione-c differs from tht of other gibberellin biosynthesis inhibitors currently in use in commercil griculture. Mny of these growth regultors, including the quternry mmonium compounds, substituted pyrimidines, norbornenodizetine derivtives, nd trizole derivtives (Grebe 1987) function by interrupting the synthesis of gibberellin erly in the biosynthetic pthwy, specificlly t the synthesis of ent-kurene. Prohexdione-C is known to interfere with the 3-ß hydroxyltion of GA 20 to GA 1. The net effect is reduction in immobile, biologiclly ctive GA 1 nd n increse in the levels of mobile, but inctive GA 20 (Evns et l. 1999, Grebe 1987). 73
74 Lo Giudice et l. Our initil interest in prohexdione-c centered on its potentil for suppressing vegettive growth of lrge, vigorous grpevines. Despite limited nd insufficient reductions in shoot growth (Lo Giudice et l. 2003), we did observe tht prohexdione-c reduced fruit set nd berry size, depending upon the timing of ppliction. These studies describe tht reproductive response. Our objectives were to quntify the response of grpes to prohexdione-c nd to determine whether reductions in berry size or crop trnslted to incresed fruit nd wine qulity ttributes. Mterils nd Methods Generl. Prohexdione-C ws pplied s Apogee (27.5% prohexdione-c) (BASF, Mount Olive, NJ) t vrible rtes in tp wter, using hnd-held, bckpck, or motorized spryer. Rtes of prohexdione-c given in tretments re s ctive ingredient. Except s indicted, ll tretments included Regulid surfctnt (polyoxyethylenepolypropoxypropnol nd lkyl 2-ethoxyethnol dihydroxy propne) (Klo, Inc., Overlnd Prk, KS) t 0.13% (v/v) nd mmonium sulfte t 900 mg/l. Ammonium sulfte hs been used to prevent the dectivtion of prohexdione-c by clcium or other ctions in hrd wter (Byers et l. 2000). Applictions were mde when there ws no rin predicted for minimum of 24 hr. Tretments were pplied to both sides of the grpevine cnopy, wetting the entire shoot including clusters to the point of runoff, unless otherwise noted. Control plots, including individul cluster tretments, were spryed with queous solutions of the djuvnts t concentrtions equl to those used in prohexdione-c tretment plots. Border plots or multiple border vines buffered tretment plots within the row, nd considertions were given to wind speed nd proximity of djcent row plots to void unintended ppliction. Vine nd pest mngement ws typicl for Virgini (Wolf nd Poling 1995). Shoot density ws djusted to 10 to 14 shoots per m of cnopy before bloom, nd primry shoots were trimmed one or two times per seson to prevent shoot tops from obstructing the fruit zone of the verticlly shootpositioned cnopies. Cbernet Suvignon, 2000. Nine-yer-old Cbernet Suvignon (Foundtion Plnt Mterils Service [FPMS] clone #7) vines, grown t the Alson H. Smith Jr. Agriculturl Reserch nd Extension Center (AHS-AREC) in Winchester, VA, were used to evlute single rte of prohexdione-c t vrible times. Vines were cordontrined nd spur-pruned on n open-lyre trellis system. Row nd vine spcing were 3.7 nd 2.1 m, respectively. The experimentl design ws completely rndomized with five replictes of three-vine plots. Tretments consisted of control nd prohexdione-c (250 mg/l) pplied either two (2xP-C) or three (3xP-C) times. The 2xP-C nd 3xP-C tretments were pplied prebloom (15 nd 31 My), while 3xP-C comprised n dditionl postbloom ppliction (30 June). Full bloom occurred 5 June. Spry volumes nd ctive ingredient pplied per plot in this nd following experiments re given in Lo Giudice (2002). Fruit ws hrvested on 20 October, t which time clusters nd totl fruit weight were recorded by vine. Individul vine dt were verged by plot for sttisticl nlyses. Fifty-berry smples were collected from ech tretment replicte immeditely before hrvest to determine verge berry weight. Berry smpling in ll experiments consisted of rndom selection of clusters, but methodicl smpling of individul berries. Berries were collected by sequentilly smpling the bck (towrd center of row) of cluster, then the front (towrd cnopy exterior) of the next cluster, the next cluster s lterl shoulder region, the next cluster s tip region, nd so on. This strtegy lessened the likelihood of disproportiontely smpling one region of cluster over nother. The opposing sides of the divided cnopy (lyretrined) vines were eqully smpled (25 berries/side). Primry shoots nd their flower clusters were counted on tretment vines in 2001, before shoot thinning, to determine if there were ny second-yer effects from prohexdione-c pplictions mde during 2000. Cbernet frnc, 2000. Four-yer-old Cbernet frnc vines, grown t Indin Springs Vineyrd ner Woodstock, VA, were used to evlute vried rtes of prohexdione-c. Vines were cordon-trined nd spur-pruned nd the cnopy ws verticlly shoot-positioned (VSP). Row nd vine spcing ws 2.7 nd 2.1 m. The experimentl design ws completely rndomized, nd ech tretment consisted of 10 single-vine replictes. Tretments were 125, 250, nd 375 mg/l prohexdione-c. To determine if mmonium sulfte itself ffected components of yield or fruit composition, two controls were included; one control included mmonium sulfte nd Regulid, while the other control included only Regulid. All vines were spryed to the point of drip once prebloom (18 My), nd twice postbloom (1 nd 30 June). Full bloom occurred 26 My. Clusters nd totl fruit weight per vine were determined t hrvest on 11 October. Fifteen-berry smples were collected from ech vine immeditely before hrvest nd were pooled to provide ech tretment with two 45-berry smples nd one 60-berry smple for sufficient smple volume. Averge berry weight ws clculted from these replictes. Primry shoots nd their flower clusters were counted on tretment vines in 2001, before shoot thinning, to determine if there were ny second-yer effects from prohexdione-c pplictions mde during 2000. Seyvl, 2000. Multiple pplictions of single rte of prohexdione-c were evluted with own-rooted, 10-yerold Seyvl (interspecific hybrid) vines, grown t the AHS- AREC. Vines were cordon-trined nd spur-pruned with VSP cnopy. Row width nd in-row vine spcing were 3.7 nd 2.1 m, respectively. The experimentl design ws completely rndomized with five replictes of three-vine plots. Tretments consisted of three pplictions of prohexdione- C (250 mg/l) nd control. The initil ppliction ws mde
Effects of Prohexdione-C on Yield nd Composition 75 prebloom on 15 My, the second on 31 My (full bloom), nd the third on 30 June. Fruit ws hrvested on 6 September. Fifty-berry smples were rndomly collected from ech of the five tretment replictes immeditely before hrvest nd frozen for lter nlysis. Clusters per vine nd totl fruit weight per vine were recorded. Clusters were visully inspected s hrvested, nd those with >5% rot were recorded s being rot ffected. No effort ws mde to chrcterize rot; however, the signs nd symptoms suggested tht most rot ws due to Botrytis cinere, to which Seyvl is susceptible. Rot severity ws lso recorded by estimting the percentge of rot-ffected berries in ech cluster nd clculting s: Rot severity = # rotten clusters/[# clusters totl x % rotten berries in ech cluster] x 100 Berry weights nd bsic fruit composition were determined from berry smples. After thwing t room temperture, smples were homogenized nd the juice ws seprted from pulp with filter bg (model 400; Stomcher Lbsystem, Sewrd Ltd., London, UK) nd centrifuged t 17,000 rpm t 10 C for 30 min. Immeditely following processing, juice soluble solids were mesured with temperture-compensted refrctometer (model 10430; Reichert Scientific Instruments, Depew, NY), ph with Fisher Accumet ph meter (model 815MP; Pittsburgh, PA), nd titrtble cidity vi procedures of Zoecklein et l. (1999). Cbernet Suvignon individul cluster experiment, 2001. Individul Cbernet Suvignon clusters were treted with prohexdione-c during 2001 to more precisely guge tretment effects on berry development t different phenologicl stges. Vines were in the sme block of Cbernet Suvignon s described bove, but hd not been previously treted with prohexdione-c. Tretments consisted of single prohexdione-c ppliction on one of 13 seprte dtes tht progressed weekly (T 1 through T 13 ) nd tht were compred with n untreted control (T 0 ). The subscripts indicte the week during which the prohexdione-c (250 mg/l) tretment ws pplied, with T 1 pplied one month fter budbrek nd T 3 pplied t 50% bloom. The Eichhorn nd Lorenz (E-L) growth stge (Coombe 1995) ws recorded t ech tretment to further define the phenologicl stge of development. Tretments were pplied to individul clusters (experimentl units). Clusters were locted on 10 vines nd ech vine constituted complete block on which ll tretments, including T 0, were included. The tretment vines were rndomly rrnged in the vineyrd nd were uniform in cnopy chrcteristics nd vine size. Tretments were rndomly ssigned to clusters on given vine so tht no tretment hd fvored position on the vine. Tretments T 0, T 2, T 5, nd T 8 were chosen for midseson destructive smpling in order to chrt the development of treted clusters. Consequently, these tretments were rndomly pplied to 20 dditionl clusters (5 per tretment vine) to provide n dequte smple size for those dtes. The experimentl units were the bsl fruit clusters of shoots tht only bore two clusters. Tretments were pplied with hnd-held spry bottle. Ech cluster ws spryed four times from different ngles to completely wet the cluster. Plstic bgs were used to isolte the trget cluster during pplictions to void sprying other clusters or leves. The first tretment (T 1 ) ws spryed one month fter budbrek t E-L 18. The four midseson smpling dtes were 21 June, 13 July, 2 August, nd 14 September. Five clusters were collected from T 0, T 2, T 5, nd T 8 on ech dte. Berries were removed from the cluster, counted, weighed, nd verge berry weight ws clculted for ech cluster. Finl hrvest of ll 10 clusters for ll tretments ws on 12 October 2001. Totl nd verge berry weight ws determined for ech cluster. Berries were then frozen, lter thwed, nd processed for fruit composition. For ech cluster, 80 berries were rndomly selected for compositionl nlyses. When clusters hd fewer thn 80 berries, ll berries from the cluster were used. After thwing t room temperture, smples were homogenized nd the juice ws seprted from pulp with filter bg nd centrifugtion, s described bove. Juice soluble solids, titrtble cidity, nd ph were mesured s described bove. Smples were then refrozen t -37 C. After thwing, 10-mL smple ws removed for juice color estimtion nd for phenol-free glycosyl-glucose (PFGG) ssy. Color ws mesured with Spectronic Genesys model-5 spectrophotometer (Thermo Spectronic US, Rochester, NY). Juice bsorbnce vlues were mesured t 280, 420, nd 520 nm. Color intensity, nthocynins, ionized nthocynins, nd totl phenols were estimted from bsorbnce vlues (Zoecklein et l. 1999). The PFGG nlysis procedures were described by Zoecklein et l. (2000), modified by Whiton nd Zoecklein (2002), nd were modifiction of previous methods to determine glycosides in juice nd wine smples by mesuring glucose (Ilnd et l. 1996, Willims et l. 1995). Briefly, must smples were rised to ph 13 to exclude phenolic glycosides, extrcted on solid-phse, nd eluted with ethnol. Elutions were hydrolyzed for 1 hr, neutrlized, nd glucose ws enzymticlly determined (Roche Dignostics, Bsel, Switzerlnd) with Thermo Lbsystems spectrophotometer (Multiskn MCC/340, Frnklin, MA). Chrdonny individul cluster experiment, 2002. Individul Chrdonny clusters were treted with prohexdione- C during 2002 to guge effects on berry development t different phenologicl stges. The experiment ws similr in purpose nd methods to the bove 2001 Cbernet Suvignon experiment. Eleven-yer-old Chrdonny (FPMS clone #4) vines grown t the AHS-AREC were used. Vine spcing nd trining were identicl to tht described for the Cbernet Suvignon. Tretments consisted of five seprte dtes (weekly) of prohexdione-c tretment (T 1 through T 5 ) nd control (T 0 ). Tretment rte, method of ppliction, repliction, nd rrngement of replictes on the vines were identicl to the Cbernet Suvignon individul cluster experiment. Tretments T 0, T 2, nd T 4 were chosen for two midseson
76 Lo Giudice et l. destructive smplings of five clusters per tretment in order to chrt the development of treted clusters. Consequently, those tretments were rndomly pplied to 10 dditionl clusters per tretment mong the test vines. The first tretment (T 1 ) ws spryed t bloom (E-L stge 23), with subsequent tretments pplied t one-week intervls. The two midseson smpling dtes were 1 July nd 1 August, t which time five clusters were collected from T 0, T 2, nd T 4 tretments. Berries were removed from the cluster, counted, weighed, nd verge berry weight clculted. Finl hrvest of 10 clusters per tretment ws on 13 September 2002, with berries per cluster nd verge berry weight obtined. Fruit soluble solids t hrvest rnged from 20.3 to 21.1 Brix. Cbernet frnc, 2001. Prohexdione-C ws pplied to Cbernet frnc in 2001 with the gol of further evluting the timing of ppliction, s well s evluting the effects on fruit nd wine composition. Five-yer-old Cbernet frnc vines, grown t Indin Springs Vineyrd nd described bove, were used. Test vines hd not been previously treted with prohexdione-c. Prohexdione-C ws pplied to the cnopy fruit zone t rte of 250 mg/l. Applictions were mde to both sides of the cnopy in zone defined, t its lowest point, by the fruiting wire nd extending verticlly 0.5 meters. Applictions were mde to the point of runoff so tht folige ws uniformly wetted. The experiment comprised two controls nd three tretments, nd the experimentl design ws completely rndomized with five replictes of four-vine plots. The three prohexdione-c tretments consisted of two prebloom pplictions (2prb) (7 nd 23 My), or single postbloom ppliction (1pstb) (19 June), or two postbloom pplictions (2pstb) (19 June nd 3 July). Bloom occurred 6 June. The E-L stges t tretment pplictions were 15 nd 17 (prebloom) nd 29 nd 31 (postbloom). Becuse prohexdione-c hd the potentil to reduce crop, ny evlution of fruit color, rom, or other qulity component hd the confounding influence of differentil crop level. We therefore included two sets of controls, one crop-thinned, the other not. On 18 My, shoots nd clusters on ll vines were counted nd thinned to 16 shoots/meter of cnopy length. In lte July, one of the controls ws cluster-thinned to n nticipted prohexdione-c induced crop level bsed on midseson cluster weights nd cluster counts for ll tretment plots. Due to the slight difference in verge cluster weight between 2prb, 1pstb, nd 2pstb, the crop per vine of the thinned control ws reduced to level intermedite to tht of the three tretments. The other control ws left with its originl cluster count. Fruit ws hrvested on 10 October nd both clusters nd totl fruit weight per vine were recorded. Fifty-berry smples were collected from ech tretment plot immeditely before hrvest. These smples were processed nd nlyzed in mnner identicl to tht described for the Cbernet Suvignon individul cluster experiment. Seprte wines were mde from ech of the Cbernet frnc tretments. Clusters were destemmed nd fter 48-hr cold sok (7 C) the must ws wrmed nd inoculted with 3.8 g/l yest (Llvin BM45, Lllemnd, Montrél, PQ). Cps were punched down three times per dy until the must fermented to dryness (2.0 g/l residul sugr) t fermenttion temperture of 65 C. The tretments were then pressed nd rcked into glss continers. Titrtble cidity, ph, ethnol, bsorbnce t 280, 420, nd 520 nm, nd phenol-free glycosides were mesured fter fermenttion. Color intensity, nthocynins, ionized nthocynins, hue, PFGG, nd totl phenols were estimted s described bove. Sensory nlysis. A tringle difference test (Lrmond 1982) ws performed on Cbernet frnc wines pproximtely four months fter completion of fermenttion. Sensory pnelists were students selected from collegite course on wine pprecition nd technology. At ech flight, pnelists (n = 29 or 30) were presented with three opque, coded glsses of wine nd told tht one differed from the other two. They were then sked to smell the wines nd indicte the different wine, fter which they were presented with three different wines for tste evlution. Tretment wines were lwys compred to untreted controls. Becuse the unthinned control hd berry weights closest to those of the tretments, the unthinned control ws chosen s the comprison wine. Dt nlysis. Dt nlysis of vrince (ANOVA) ws conducted with GLM or t test functions of SAS (version 8.2; SAS Institute, Inc., Cry, NC), nd mens compred with either single degree-of-freedom contrsts or Duncn s multiple rnge test (DMRT). Individul vine dt from multivine plots were verged by plot prior to ANOVA. Correltions between berry weight nd fruit composition in the Cbernet Suvignon individul cluster experiment were lso evluted with SAS procedures. Sttisticl nlyses were not performed on wine chemistry dt becuse wine lots were not replicted. Men vlues for prefermenttion fruit chemistry nlyses were clculted from the five tretment replictes, nlyzed with ANOVA procedures, nd mens seprted with DMRT. Tringle test results were nlyzed with rpid nlysis tble prepred by Roessler et l. (1948). Results Cbernet Suvignon, 2000. Averge cluster weight, berries per cluster, berry weight, crop per vine, nd clusters per shoot in the subsequent seson were ll decresed by pplictions of 250 mg/l prohexdione-c (Tble 1). Berries per cluster nd crop per vine were both greter for the 3xP- C tretment thn for the 2xP-C tretment. Reductions in berries per cluster were due to bortion nd withering of individul florets. Cbernet frnc, 2000. There were few differences noted between the two controls, one exception being slight decrese in berry weight with the +N control compred to the -N control (Tble 2). Becuse of these smll differences,
Effects of Prohexdione-C on Yield nd Composition 77 prohexdione-c tretments were compred to the control tht included mmonium sulfte. Prohexdione-C reduced clusters per vine, cluster weight, berries per cluster, berry weight, crop per vine, nd clusters per shoot in the subsequent seson. A negtive liner reltionship ws observed between prohexdione-c concentrtion nd clusters per vine, berries per cluster, berry weight, nd clusters per shoot, while qudrtic reltionship ws observed for cluster weight nd crop per vine. Agin, the reduction in berries per cluster ppered due to senescence of individul florets; however, the reductions in clusters per vine were due to senescence of the entire cluster similr to tht described for inflorescence necrosis. Seyvl, 2000. Cluster weight, berries per cluster, berry weight, nd crop weight per vine were ll decresed by prohexdione-c (Tble 3). The rot severity of fruit from prohexdione-c-treted vines (3.5%) ws comprble to tht of control vines (5.6%). No differences in juice soluble solids (19.5 Brix for control versus 18.6 Brix for prohexdione-c) or ph (3.3 for both) were observed. Seyvl berry skins, normlly greenish-gold, cquired reddish cst on vines treted with prohexdione-c. Cbernet Suvignon individul cluster experiment, 2001. Prohexdione- C pplictions before nd during bloom (T 1-3 ) severely reduced fruit set (Tble 4 nd Figure 1). The miniml vrition in berry weight observed for T 2 nd T 3 reflects the lmost complete loss of fruit set with clusters of those two tretments. Clusters from T 2 nd T 3 pplictions were so severely ffected it ws impossible to gther ny fruit chemistry results. Prohexdione-C cesed to ffect fruit set beyond T 3 (E-L stge 26); in fct, T 4 hd more berries per cluster t hrvest thn ny other tretment, including the control (Tble 4). Vines t T 4 responded to tretment with 51% reduction in berry weight. A reduction in berry weight ws evident for pplictions through T 6, lthough this response ws less intense fter T 4 (Tble 4). Tretments T 4 nd T 5 (E-L stges 26 nd 27) were most effective in reducing berry weight without dversely ffecting berry set; however, there ws lso certin mount of vrition in berry Tble 1 Cbernet Suvignon yield components in 2000 nd flower clusters per shoot counted in spring of 2001, in response to two (2xP-C) or three (3xP-C) pplictions of prohexdione-c t 250 mg/l in 2000. 2000 seson mesures Clusters/ Cluster wt Berries/ Berry wt Crop/ Clusters/ Tretment vine (g) cluster (g) vine (kg) shoot 2001 Control 72.7 127.03 92 1.38 9.24 1.5 2xP-C 75.3 36.11 46 0.83 2.75 1.4 3xP-C 76.1 48.41 58 0.81 3.71 1.4 Significnce of contrst (p vlue) Control vs 0.243 <0.001 <0.001 <0.001 <0.001 0.003 men of tretments 2xP-C vs 0.781 0.063 0.046 0.523 0.024 0.389 3xP-C Single degree-of-freedom contrsts between specified tretments. Tble 2 Cbernet frnc yield components in 2000 nd flower clusters per shoot counted in spring of 2001, in response to three rtes of prohexdione-c (125, 250, nd 375 mg/l) nd two controls, either with (+N) or without (-N) mmonium sulfte, in 2000. 2000 seson mesures Clusters/ Cluster wt Berries/ Berry wt Crop/ Clusters/ Tretment vine (g) cluster (g) vine (kg) shoot 2001 -N 53.5 100.8 58 1.74 5.4 1.6 +N 46.8 105.4 69 1.55 4.9 1.6 125 mg/l 34.6 68.0 48 1.44 2.4 1.3 250 mg/l 31.0 42.5 38 1.11 1.4 1.3 375 mg/l 19.1 25.4 26 0.95 0.5 1.1 Significnce of contrst (p vlue) +N vs -N 0.054 0.530 0.083 0.001 0.173 0.711 +N vs <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 men of tretments Liner <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Qudrtic 0.950 0.052 0.295 0.544 0.001 0.103 Single degree-of-freedom contrsts nd trend nlysis of specified tretments. Tble 3 Seyvl yield components s ffected by three pplictions of prohexdione-c (P-C) in 2000. Clusters/ Cluster wt Berries/ Berry wt Crop/vine Tretment vine (g) cluster (g) (kg) Control 48.8 198.0 79 2.53 9.8 P-C 45.5 131.8 63 2.10 6.0 Probbility > t t test ns 0.004 0.047 <0.001 0.020 ns: not significnt.
78 Lo Giudice et l. weight with those smples, s illustrted in Figure 1. Berry development of the control (T 0) nd T 8 vines ws nerly identicl over the four smpling dtes (Figure 2), with T 5 nd T2 exhibiting correspondingly lower terminl berry sizes nd development curves. The immedite postbloom ppliction (T4) hd no impct on soluble solids (Brix), ph, or titrtble cidity (Tble 5); however, the T 4 tretment did increse phenols, totl nd ionized nthocynins, nd color intensity (420 + 520 nm). Phenol-free glycoside concentrtion ws highest for the T4 tretment nd similr mong other tretments. Berry weight ws inversely nd significntly correlted to juice spectrl qulity. The correltion coefficients between bsorbnce t 280, 420, nd 520 nm nd berry weight, cross ll tretments, rnged from -0.55 to -0.58, ll highly significnt. The strength of those correltions incresed to -0.67 to -0.77, still highly significnt if exmined for T4 tretment only. Prohexdione-C ws ineffective in reducing berry weight when pplied fter T6 (Tble 4). Accompnying this return to norml berry weight ws return to bsorbnce nd PFGG vlues tht were comprble to those of the control (Tble 5). Tble 4 Eichhorn nd Lorenz (E-L) stge t tretment, berries per cluster, nd berry weight of Cbernet Suvignon nd Chrdonny clusters treted with 250 mg/l prohexdione-c. Bloom (50% cp-fll, or E-L stge 23) is T 3 for Cbernet Suvignon nd T1 for Chrdonny. Cbernet Suvignon (2001) Tretment E-L stge Berries/ cluster Berry wt (g) T0 T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 T12 T13 n b 18 21 23 26 27 29 32 33 33 33 33 34 37 127.5 bc c 44.1 d 0.2 e 4.5 e 166.2 132.0 bc 133.4 bc 136.4 b 108.1 bc 128.6 bc 131.7 bc 98.3 c 108.8 bc 98.0 c 1.42 1.25 0.86 0.32 0.69 1.09 1.30 1.47 1.48 1.48 1.48 1.52 1.46 1.46 bc c e g f d bc b b b b b b Chrdonny (2002) E-L stge n 23 25 29 32 33 Berries/ cluster 198.5 89.7 139.1 173.7 143.0 150.7 c b b b b Berry wt (g) 1.68 1.08 1.22 1.33 1.62 1.79 c bc b Subscripts indicte week t which tretment ws pplied. n: not pplicble. cmens within columns followed by the sme letter do not differ t p 0.05. b Chrdonny individul cluster experiment, 2002. The bloom (T 3) prohexdione-c ppliction significntly reduced both berry weight nd berries per cluster (Tble 4), while berry weight, but not berry set, continued to be negtively ffected by tretments mde t T2 (E-L 25) nd T3 (E-L 29). The rtes of berry development over time ppered similr mong the control, T2, nd T4 tretments (dt not shown); however, T 2 hd consistently smller berries thn the other two tretments. Cbernet frnc, 2001. Although the cropthinned vines hd significntly fewer clusters thn did the unthinned tretment, tht did not ffect cluster weight or crop per vine (Tble 6). Due to the lck of significnt differences between the controls, tretments were compred to unthinned tretment. Only the 2pstb tretment differed significntly from either of the controls in terms of components of yield. Two postbloom prohexdione-c pplictions reduced crop per vine without ny significnt reduction in cluster weight or berries per cluster. The reduction in crop ws principlly due to reduction in clusters. Prohexdione-C did not ffect fruit soluble solids, but ll tretments incresed ph (Tble 7). Figure 1 Cbernet Suvignon berry weight t hrvest s ffected by single pplictions of 250 mg/l prohexdione-c to individul clusters t weekly (T vlue) intervls in 2001. Box whiskers indicte extreme vlues, the box bottom nd top define the first nd third dt qurtiles, the cross br indictes the medin, nd + indictes the smple men. T0 is the untreted control, T1 ws pplied one month fter budbrek, nd T3 corresponds to 50% bloom. Figure 2 Averge berry weights of 2001 Cbernet Suvignon clusters treted with 250 mg/l prohexdione-c. T numbers indicte week of tretment ppliction, followed by Eichhorn-Lorenz (E-L) stge t tretment. T0 is the untreted control.
Effects of Prohexdione-C on Yield nd Composition 79 Tble 5 Must soluble solids (Brix), ph, titrtble cidity (TA), hue, totl nthocynins (nth), ionized nthocynins, totl phenols, color intensity, nd phenol-free glycosyl-glucose (PFGG) for Cbernet Suvignon individul clusters to which pplictions of 250 mg/l prohexdione-clcium were mde in 2001. Hue Ionized Totl Intensity TA (420/520 Anth nth phenols (420+520 PFGG Tretment Brix ph (g/l) nm) (mg/l) (mg/l) (u) nm) (µm) T 0 21.1-d 3.87 5.4-d 0.74b 95bc 16.5c 4.8bc 1.6bc 289.2bc T 1 21.5b 3.73b 5.4b-d 0.70b 103-c 20.5b 6.2b 2.0b 318.3bc T 2 n b n n n n n n n n T 3 n n n n n n n n n T 4 21.0-d 3.72b 6.1 0.69b 139 22.8 10.3 3.2 542.6 T 5 21.6 3.82b 5.6-d 0.75b 109b 17.8bc 6.2b 2.1b 333.4b T 6 20.6c-e 3.78b 5.8b 0.78b 82bc 17.8bc 4.5bc 1.5bc 347.1b T 7 20.5c-e 3.79b 5.5-d 0.70b 83bc 18.0bc 3.6bc 1.5bc 219.3bc T 8 20.2e-f 3.77b 5.6-c 0.77b 80bc 16.4c 3.5bc 1.3bc 242.1bc T 9 20.0ef 3.81b 5.5-d 0.75b 83bc 15.0c 3.6bc 1.3bc 334.2b T 10 20.9-d 3.79b 5.3b-d 0.78b 72bc 14.4c 3.1bc 1.3bc 299.1bc T 11 19.6f 3.80b 5.1cd 0.78b 69c 16.5c 3.2bc 1.1c 284.9bc T 12 20.4c-e 3.86 5.1cd 0.80b 64c 15.8c 2.7c 1.1c 179.4c T 13 20.7b-e 3.85 4.9d 0.84 70bc 16.2c 3.3bc 1.1c 292.5bc Mens within columns followed by the sme letter do not differ t p 0.05. b n: dt not vilble. Tble 6 Components of yield for Cbernet frnc treted with two prebloom (2prb), one postbloom (1pstb), or two postbloom (2pstb) pplictions of 250 mg/l prohexdione-c during the 2001 seson, compred to n unthinned (UT) nd thinned (T) control. Clusters/ Cluster Berries/ Berry Crop/vine Tretment vine wt (g) cluster wt (g) (kg) UT 54.5 77.35 52.1 1.49 4.24 T 39.4 93.27 62.2 1.50 3.72 2prb 52.8 69.44 49.5 1.40 3.67 1pstb 53.3 69.81 51.0 1.38 3.72 2pstb 47.1 61.68 44.6 1.39 2.94 Significnce of contrst (p vlue) UT vs T <0.001 0.078 0.075 0.864 0.344 UT vs 2prb 0.568 0.367 0.635 0.178 0.303 UT vs 1pstb 0.695 0.390 0.843 0.103 0.342 UT vs 2pstb 0.023 0.082 0.184 0.128 0.025 Single degree-of-freedom contrsts of specified tretments. Titrtble cidity ws generlly not ffected with the exception of decrese cused by the two postbloom pplictions. Two prebloom pplictions or single postbloom ppliction incresed totl nthocynins, totl phenols, nd color intensity. Aside from ph nd TA, fruit chemistry for 2pstb ws not ffected by tretment. There were no significnt differences for PFGG with ny tretment. Wine nlyses were similr mong tretments nd controls for ethnol, titrtble cidity, nd ph (Tble 8). Prohexdione-C incresed wine totl nthocynins, totl phenols, nd color intensity, prticulrly with the postbloom pplictions. Lcking repliction, however, it is uncertin whether those differences were significnt. The wine chemistry dt re somewht consistent with the must dt (Tble 7), one difference being the reltive differences between the single nd double postbloom Tble 7 Cbernet frnc must soluble solids (Brix), ph, titrtble cidity (TA), hue, totl nthocynins (nth), ionized nthocynins, totl phenols, color intensity, nd phenol-free glycosyl-glucose (PFGG) s ffected by two prebloom (2prb), one postbloom (1pstb), or two postbloom (2pstb) pplictions of 250 mg/l prohexdione-clcium during the 2001 seson, compred to n unthinned (UT) nd thinned (T) control. Hue Ionized Totl Intensity TA (420/520 Anth nth phenols (420+520 PFGG Tretment Brix ph (g/l) nm) (mg/l) (mg/l) (u) nm) (µm) UT 22.3 3.44 5.35 0.85 108 11.9 17.9 5.5 298 T 22.6 3.51 5.50 0.97 107 4.3 17.5 5.2 577 2prb 23.0 3.49 5.38 0.76 153 4.1 21.6 6.9 289 1pstb 22.5 3.50 5.10 0.68 179 4.1 22.4 7.6 340 2pstb 22.6 3.51 4.86 0.84 113 4.0 18.8 5.6 233 Significnce of contrst (p vlue) UT vs T 0.212 0.035 0.425 0.428 0.976 0.134 0.836 0.634 0.190 UT vs 2prb 0.368 0.046 0.867 0.170 0.057 0.123 0.059 0.038 0.967 UT vs 1pstb 0.469 0.023 0.195 0.070 0.005 0.121 0.025 0.003 0.847 UT vs 2pstb 0.337 0.007 0.018 0.415 0.830 0.118 0.636 0.884 0.764 Single degree-of-freedom contrsts of specified tretments.
80 Lo Giudice et l. Tble 8 Cbernet frnc wine lcohol (ETOH), titrtble cidity (TA), ph, totl nthocynins (nth), ionized nthocynins, totl phenols, color intensity nd phenol-free glycosyl-glucose (PFGG) s ffected by prohexdione-c tretments in 2001. Tretments included unthinned (UT) nd crop-thinned (T) controls, nd prohexdione-c pplied twice prebloom (2prb), once postbloom (1pstb) or twice postbloom (2pstb) t 250 mg/l. Ionized Totl Intensity ETOH TA Anth nth phenols (420+520 PFGG Tretment (%) (g/l) ph (mg/l) (mg/l) (u) nm) (µm) UT 11.30 7.48 3.40 377 20.7 28.6 6.3 104 T 11.95 7.63 3.47 390 22.5 31.3 7.1 110 2prb 11.70 7.50 3.43 384 22.8 30.3 6.9 101 1pstb 11.80 7.44 3.43 424 24.7 33.7 8.4 102 2pstb 12.00 7.68 3.44 473 37.6 40.8 13.4 113 Tble 9 Cbernet frnc tringle test of wine rom nd flvor compring n unthinned control (UT) with either two prebloom pplictions (2prb), one postbloom ppliction (1pstb), or two postbloom pplictions (2pstb) of 250 mg/l prohexdione-c in 2001. Pnelists Correct Comprison (n) responses p vlue UT vs 2prb Arom 30 9 >0.40 UT vs 2prb Flvor 30 13 0.20 UT vs 1pstb Flvor 29 11 >0.40 UT vs 1pstb Tste 29 9 >0.40 UT vs 2pstb Arom 29 12 0.30 UT vs 2pstb Flvor 29 12 0.30 pplictions. Despite fruit nd potentil wine chemistry differences between the tretments, sensory pnels could not distinguish differences mong the wines in tringle tests (Tble 9). Discussion Prohexdione-C hd the potentil to severely reduce fruit set nd crop yield. This ws prticulrly evident with prebloom (E-L stges 18 to 21) nd bloom (E-L stge 23) pplictions. Reduced set ws observed in V. vinifer cultivrs s well s the interspecific cultivr Seyvl. Reduced set ppered to result from destruction of developing flowers; the prebloom tretments cused browning nd bscission of flowers nd, t higher rtes, bscission of entire clusters t the peduncle. No corresponding necrosis or other lesions were observed on folige or other tissues, indicting tht the effects of prohexdione-c were loclized in reproductive tissues. Prohexdione-C inhibits the conversion of minocyclopropnecrboxylic cid into ethylene, nd its ppliction reduces ethylene-induced senescence in severl plnt systems (Rdemcher et l. 1992). The senescence nd bscission of grpevine reproductive structures, therefore, is not esily explined s response to reduced ethylene. The only prebloom pplictions tht filed to reduce set were the two prebloom pplictions mde to Cbernet frnc in 2001, t E-L stges 15 nd 17. Why those tretments filed to reduce set is uncertin. The prebloom tretment of Cbernet frnc in 2000 ws bout one week before bloom nd set ws reduced. The second of the two prebloom tretments to Cbernet frnc in 2001 ws bout two weeks before bloom. Thus, it might be tht the Cbernet frnc clusters were not yet sensitive to the prohexdione-c in the 2001 experiment. Reduction in fruit set my hve two benefits. One potentil benefit would be purposeful reduction in crop to increse fruit nd wine qulity. Becuse crop thinning typiclly is done mnully, it is expensive. Hence, chemicl mens of thinning could sve money. The lck of berry set reductions with the 2001 Cbernet frnc experiment suggests, however, tht timing of prohexdione-c ppliction is criticl to chieve predictble thinning effect. A second potentil benefit of reduced set might be relized s reduction in bunch rots with rot-prone cultivrs. We did not observe such reduction during the course of these experiments; however, the overll incidence of rot ws very low. Further investigtions of the effects of prohexdione-c on berry set, berry size, nd bunch rot susceptibility re wrrnted under more disese-conducive conditions. The single-cluster Cbernet Suvignon nd Chrdonny experiments provided more precise mesure of when the clusters were sensitive to prohexdione-c ppliction. The most significnt effects were reduced fruit set for pplictions mde before nd during bloom, nd reduced berry weight for the immedite postbloom pplictions. Vrition in berries per cluster nd berry weight might hve been due to uneven distribution of spry mteril on the clusters, synchrony of flower nd berry development on given cluster, or combintion of those fctors. Becuse the single-cluster experiments trgeted clusters, it is pprent tht the clusters cn directly bsorb the prohexdione-c nd tht the effect is not medited by folir bsorption. The severity of fruit-set reductions tht we observed with prebloom nd bloom pplictions ws unexpected. The literture describing the effect of other gibberellin biosynthesis inhibitors on fruit set is conflicting, with some reserchers noting decreses in fruit set (Reynolds et l. 1992), some seeing no impct (Hunter 1992), nd others citing increses (Coombe 1965). To our knowledge, this is the first forml report of the effects of prohexdione-c on grpe berry development, nd informtion on the reproductive response with other fruits is meger. Prohexdione-C pplictions suppressed vegettive growth of pple tree shoots but hd no effect on fruit, other thn slight increse in red colortion, which ws ttributed to incresed light interception (Byers nd Yoder 1999).
Effects of Prohexdione-C on Yield nd Composition 81 Only recently hs the chnge in concentrtion of bioctive gibberellins during the course of berry development been studied. New literture describes shrp increse in GA 3 levels 21 nd 65 dys fter nthesis in seeded cultivrs of V. vinifer (Perez et l. 2000). The reductions in berry weight observed with the immedite postbloom pplictions of prohexdione-c my reflect depression of bioctive GA 3 in berries. The 2000 Cbernet frnc experiment illustrted tht yield components declined linerly with incresing concentrtion of prohexdione-c. Ammonium sulfte lone might hve cused some decrese in berry weight; however, other components of yield were unffected by the ddition of this wter conditioner. In light of the improved response of pple to prohexdione-c when mmonium sulfte ws dded to the spry solution (Byers et l. 2000), we felt tht it ws prudent to use the conditioner until further evlutions could be conducted on grpe. The 375 mg/l tretment cused prticulrly severe crop reduction with Cbernet frnc in 2000. Surprisingly, 250 mg/l rte of prohexdione-c ppliction did not reduce Cbernet frnc yield components in 2001. A possible reson for this discrepncy is tht the vines in the 2000 experiment received three pplictions of prohexdione-c, compred to only two in 2001. However, considering the lck of difference between two nd three pplictions to Cbernet Suvignon in 2000, it is likely tht the third ppliction to Cbernet frnc hd little effect. The inconsistency of effects between yers my lso relte to the fct tht the timing of the prebloom pplictions differed slightly between the two sesons. A further possibility is tht environmentl conditions (such s temperture nd sunlight conditions) differed between sesons nd medited the tretment effects. Prohexdione-C unexpectedly depressed subsequent seson shoot fruitfulness. The clusters of given seson originte from nlgen of the previous seson tht differentited s either tendril or inflorescence primordi (Mullins 1986). Cytokinins fvor the differentition of nlgen into cluster primordi nd regulte the formtion of individul florets on the developing cluster primordi in the yer of nthesis (Mullins 1986). Mny of the nitrogen-contining, heterocyclic growth regultors cuse elevted levels of cytokinins in plnt tissues (Rdemcher 2000), so one might suspect tht prohexdione-c would led to incresed cytokinins nd possibly incresed fruitfulness. In contrst, gibberellin pplictions during the differentition of nlgen cuse nlgen to form tendril primordi rther thn inflorescence primordi (Srinivsn nd Mullins 1980). Agin, one might therefore expect tht the ppliction of gibberellin biosynthesis inhibitor to the differentiting nlgen in 2000 would hve incresed the cluster number in 2001. We do not hve n explntion for the reduced fruitfulness observed in the second seson. Prohexdione-C hd miniml effects on fruit soluble solids, ph, or titrtble cidity. Direct effects of prohexdione- C on strch, sugr, nd color in pple lso seem to be miniml or lcking (Byers nd Yoder 1999). In contrst to the miniml effects on fruit primry metbolites, prohexdione-c did pper to hve some effects on the concentrtions of some secondry metbolites. For exmple, incresed levels of PFGGs were significntly nd negtively correlted with berry weight in the Cbernet Suvignon individul cluster experiment. Others hve found sunlight exposure to ffect fruit qulity. For instnce, lef thinning to increse fruit-zone light exposure incresed totl glycosides 17 to 29% in Riesling (Zoecklein et l. 1998). Bureu et l. (2000) found tht light exposure could increse the concentrtion of C 13 -norisoprenoids pproximtely 40%. We believe, however, tht the incresed concentrtion of PFGG is more likely due to reduced berry size thn to incresed sunlight exposure, becuse we were unble to lter the light environment of grpevine cnopies with prohexdione-c pplictions (Lo Giudice et l. 2003). Our results suggest tht nthocynins might lso be incresed by prohexdione-c, nd two possible mechnisms re offered. The more obvious explntion would be n increse in surfce to volume rtio, resulting in n increse in nthocynins on per berry bsis. A second possibility is tht nthocynin production is directly ffected by ltertions in the flvonoid biosynthetic pthwy. While not mesured, we did note n incresed red colortion of Seyvl (but not Chrdonny) berries tht hd been treted with prohexdione-c. Countering the direct effect hypothesis re reports tht prohexdione-c nd relted growth retrdnts inhibit the formtion of nthocynins in some other crops (Rdemcher et l. 1992, 2000). The bsis of tht inhibition ppers to be impirment of flvnone 3- hydroxylse (FHX), key enzyme in the synthesis of dihydroquercetin, leucocynidin, nd other flvonoids (Rdemcher 2000). The inhibition of FHX in pple (Mlus sp.) leds to buildup of the novel flvonoids, luteoliflvn nd eriodictyol (Roemmelt et l. 1999). Detiled nlyses of the biosynthetic ctivity of prohexdione-c in grpe hve yet to be pursued. Unlike the individul cluster tretments used with Cbernet Suvignon, the color response of Cbernet frnc must to prohexdione-c ppered unrelted to berry weight. For exmple, the two prebloom nd the single postbloom pplictions were ssocited with incresed bsorbnce vlues nd incresed totl phenols, even though the berry size ws not significntly reduced. Cbernet frnc wine color ppered to be gretest for the two postbloom pplictions. These results my be due to the existence of direct tretment effect independent of berry weight, s discussed bove. It is lso possible tht slight vritions in fermenttion conditions ffected the wine chemistry results. The use of replicted winemking lots in the future my help identify whether the observed effects re reproducible. Despite some differences in juice nd wine composition ssocited with prohexdione-c tretments, the sensory
82 Lo Giudice et l. pnelists could not distinguish Cbernet frnc control from tretment wines. Wines were very young when judged nd subtle differences my not hve emerged or been perceptible to n untrined pnel. Conclusion Prohexdione-C reduced fruit set nd berry weight in rte- nd timing-dependent fshion. With the exception of Cbernet frnc in 2001, prebloom nd bloom pplictions hd prticulrly depressive effect on fruit set, while immedite postbloom pplictions hd less effect on fruit set but more depressive effect on berry size. Prohexdione-C my therefore provide useful mens of reducing berry size or reducing fruit set where those responses re desired. The reductions in berry weight corresponded to higher phenol-free glycosyl-glucose (PFGG) concentrtions, greter totl phenols, nd greter color intensity of Cbernet Suvignon must. Similr responses were mesured with Cbernet frnc musts, lthough PFGG concentrtions were insignificntly ffected. Mny rom nd flvor precursors re glycosidiclly bound in must nd young wine. Although n increse in must PFGG might led to improved wine sensory chrcteristics, we were unble to demonstrte sensory improvements from prohexdione-c with our pnelists in young Cbernet frnc wines. Becuse Cbernet frnc fruit nd wine chemistry results were independent of berry weight, the possibility of direct tretment effect, unrelted to berry size, is plusible. Literture Cited Bureu, S.M., R.L. Bumes, nd A.J. Rxungles. 2000. Effects of vine or bunch shding on the glycosylted flvor precursors in grpes of Vitis vinifer L. cv. Syrh. J. Agric. Food Chem. 48:1290-1297. Byers, R.E., nd K.S. Yoder. 1999. Prohexdione-clcium inhibits pple, but not pech, tree growth, but hs little influence on pple fruit thinning or qulity. HortScience 34:1205-1209. Byers, R.E., D.H. Crbugh, nd L.D. Combs. 2000. The influence of Apogee nd its combintions with ethephon, chemicl thinners, ctions, nd/or djuvnts for pple tree growth control nd return bloom. In Proceedings of the 27 th Annul Meeting of the Plnt Growth Regultion Society of Americ, pp. 187-192. PGRSA, LGrnge, GA. Coombe, B.G. 1965. Increse in fruit set of Vitis vinifer by tretment with growth retrdnts. Nture 205:305-306. Coombe, B.G. 1992. Reserch on development nd ripening of the grpe berry. Am. J. Enol. Vitic. 43:101-110. Coombe, B.G. 1995. Adoption of system for identifying grpevine growth stges. Aust. J. Grpe Wine Res. 1:100-110. Coombe, B.G., nd M.G. McCrthy. 2000. Dynmics of grpe berry growth nd physiology of ripening. Aust. J. Grpe Wine Res. 6:131-135. Estebn, M.A., M.J. Villnuev, nd J.R. Lissrrque. 1999. Effect of irrigtion on chnges in berry composition of Temprnillo during mturtion. Am. J. Enol. Vitic. 50:418-434. Evns, J.R., R.R. Evns, nd C.L. Regusci. 1999. Mode of ction, metbolism, nd uptke of BAS 125W prohexdione-clcium. HortScience 34:1200-1201. Fernndez de Simon, B., T. Hernndez, nd I. Estrell. 1993. Phenolic composition of white grpes (vr. Airen). Chnges during ripening. Food Chem. 47(1):47-52. Grebe, J.E. 1987. Gibberellin biosynthesis nd control. A review. Ann. Rev. Plnt Physiol. 38:419-465. Hunter, D.M. 1992. Pclobutrzol ffects growth nd fruit composition of potted grpevines. HortScience 27:319-321. Ilnd, P.G., W. Cynkr, I.L. Frncis, P.J. Willims, nd B.G. Coombe. 1996. Optimistion of methods for the determintion of totl nd red-free glycosyl glucose in blck grpe berries of Vitis vinifer. Aust. J. Grpe Wine Res. 2:171-178. Lrmond, E. 1982. Lbortory Methods for Sensory Evlution of Food. Publiction 1637. Cnd Deprtment of Agriculture, Ottw, ON. Lo Giudice, D. 2002. The impct of prohexdione-clcium on grpe vegettive nd reproductive development nd wine chemistry. Thesis, Virgini Polytechnic Institute nd Stte University, Blcksburg, VA. Lo Giudice, D., T.K. Wolf, nd R.P. Mrini. 2003. Vegettive response of Vitis vinifer to prohexdione-clcium. HortScience 38:1435-1438. Mullins, M.G. 1986. Hormonl regultion of flowering nd fruit set in the grpevine. Act Hortic. 179:309-315. Owens, C.L., nd E. Stover. 1999. Vegettive growth nd flowering of young pple trees in response to Prohexdione-clcium. HortScience 34:1194-1196. Prk, S.K., nd A.C. Noble. 1993. Monoterpenes nd monoterpene glycosides in wine roms. In Beer nd Wine Production: Anlysis, Chrcteriztion, nd Technologicl Advnces. B.H. Gump nd D.J. Pruett (Eds.), pp. 98-109. Americn Chemicl Society, Wshington, DC. Perez, F.J., C. Vini, nd J. Retmles. 2000. Bioctive gibberellins in seeded nd seedless grpes: Identifiction nd chnges in content during berry development. Am. J. Enol. Vitic. 51:315-318. Rdemcher, W. 2000. Growth retrdnts: Effects on gibberellin biosynthesis nd other metbolic pthwys. Ann. Rev. Plnt Physiol. Plnt Mol. Biol. 51:501-531. Rdemcher, W., K.E. Temple-Smith, D.L. Griggs, nd P. Hedden. 1992. The mode of ction of cylcyclohexnediones: A new type of growth retrdnt. In Progress in Plnt Growth Regultion. C.M. Krssen et l. (Eds.), pp. 571-577. Kluwer Acdemic, Dordrecht, The Netherlnds. Reynolds, A.G., nd A.P. Nylor. 1994. Pinot Noir nd Riesling grpevines respond to wter stress durtion nd wter holding cpcity. HortScience 29:1505-1510. Reynolds, A.G., D.A. Wrdle, A.C. Cottrell, nd A.P. Gunce. 1992. Advncement of Riesling fruit mturity by Pclobutrzol-induced reduction of lterl shoot growth. J. Am. Soc. Hortic. Sci. 117:430-435. Roemmelt, S., D. Treutter, J.B. Spekmn, nd W. Rdemcher. 1999. Effects of prohexdione-c on the flvonoid metbolism of pple with respect to plnt resistnce ginst fire blight. Act Hortic. (ISHS) 489:359-364. Roessler, E.B., J. Wrren, nd J.F. Guymon. 1948. Significnce in tringulr tste tests. Food Res. 13:503-505.