» » *» » >«■ ■» l» .»»««»_■». Please handle this volume with care. The University of Connecticut Libraries, Storrs 3 =1153 DllMflTflE y GAYLORD RG Digitized by the Internet Archive in 2011 with funding from LYRASIS members and Sloan Foundation http://www.archive.org/details/winegrapetrials920kiyo Co/mM s H-3> .EA2. The 1 no^ap Connecticut g Agricultural | Wine Grape Experiment s Trials Station, 5 1990-1993 New Haven 5 BY RICHARD K. KIYOMOTO Bulletin 920 April 1994 SUMMARY Since the passage of the Farm Winery Bill in 1978, Connecticut has seen the acreage devoted to grapes increase from fewer than 20 acres to 325 by 1989. However, Connecticut wineries must still rely on neighboring viticultural districts to meet their demand for wine grapes. A 4-year study was undertaken to follow the yield of 1 8 grape varieties in an unreplicated experimental planting in Hamden, CT in order to provide information on performance of wine grapes in Connecticut. Data from 1990-1993 show excellent performance by two promising French hybrid cultivars which are not currently grown in Connecticut on a commercial scale. The red wine cultivar Chambourcin and the white wine cultivar Villard Blanc showed yield, sugar, and acid characteristics which compared favorably with the white wine cultivar Seyval which has a proven record of hardiness and productivity in Connecticut. Wine Grape Trials 1990-1993 BY RICHARD K. KIYOMOTO Since the passage of the Farm Winery Bill in 1978, the acreage of grapes in Connecticut increased from fewer than 20 acres to 325 by 1 989, and the number of farm wineries increased from one to eight during the same period (Baxevanis, 1989). Despite the increase in acreage devoted to wine grapes, the supply of Connecticut-grown grapes is inadequate. In response to the increased interest in grapes, the Experiment Station began experiments in 1978 to investigate the winter hardiness and disease susceptibility of both table and wine grapes (Walton, 1987). The original plantings were made in 1978 and 1979. Additional cultivars were added during 1982-1987 after others were discarded for poor per- formance. Pesticides were applied sparingly in order to assess disease susceptibility. Although these early trials included both table and wine grape cultivars, I have focused on the performance of the wine grapes and report vine vigor, sugar content, and acidity of the grapes. MATERIALS AND METHODS Table 1 lists the wine grape cultivars in the trial and the date of planting at Lockwood Farm in Hamden, CT. Vines were trained with single trunks to an umbrella Kniffin sys- tem with three to four canes approximately 36 inches in length each. Trellis wires were spaced 3 and 6 ft above the ground. Five vines of each cultivar were planted within a row at 6 ft spacing. Each row consisted of 10 vines of two cultivars. There were two plantings of Seyval and Cham- bourcin. Rows were spaced 9 ft apart with a mixed grass groundcover. A weed-free strip (2 ft wide) was maintained down the planted row around the vines. Fertilizer, 10-10-10, was applied as a split application in May and early July to supply a total of 40 lbs each of N, P2O5, and K2O. To encourage air movement beneath the canopy, I removed leaves from around developing bunches in June and pruned canes in June and July so the canopy never draped to within 3 ft of the soil. Depending on the berry set, soil moisture, and vine vigor, bunches were thinned to two per shoot in 1990, Table 1 . Wine grapes in trial. Cultivar Year Planted Juice Color Hybrid and Non-Vinifera Grapes: Cayuga White 1983 White Horizon 1985 White Seibel 10868 1985 White Seyval 1980 White Verdelet 1985 White Vidal 256 1986 White Vignoles (Ravat 51) 1983 White Villard Blanc 1985 White Chambourcin 1984 Red Chancellor 1984 Red Chelois 1986 Red DeChaunac 1983 Red S26-627 1986 Pink Villard Noir 1985 Red Vinifera Grapes: Chardonnay 1984 White Gewurztraminer 1984 White Riesling 1984 White Gamay Beaujolais 1986 Red one to two per shoot in 1991 and 1992, and one per shoot in 1993. In early July, 1993 only one bunch was left per shoot to reduce the extreme drought stress on the vines. At the start of these studies in the winter 1989-1990, cultivars showed great differences in the amount of the canes produced, which was taken to be an indicator of vigor. Vig- orous vines produced large quantities of canes which re- quired pruning. Visual ratings were made on the amount of cane material collected from each cultivar during winter pruning in 1989-1990 and 1990-1991 (0-3 scale with 0 representing no prunings and 3 representing the largest Connecticut Agricultural Experiment Station Bulletin 920 amount of pruned wood). Fruit set was evaluated in the springs of 1990 and 1991 (0-5 scale with 0 representing no fruit set and 5 representing the largest amount of fruit set). All fruit were removed from cultivars with poor cane vigor in spring 1990 to allow the vines to recover. Vine renovation was attempted by pruning most of the old wood and training new vines from strong basal shoots. No insecticides were applied from 1990-1993, and fungi- cide applications were adjusted to rainfall and drying condi- tions. Figure 1 plots the cumulative rainfall from Julian Date 126-250 (10 June-7 September) for the 4 years of study. Julian Dates number the days of the year sequentially begin- ning with January 1 as Julian Day 1 . Three fungicide applica- tions were made in 1990 and 1991. Fungicides were applied four times in 1992, a season with many cool, moist days. In 1993, a dry spring and summer, fungicides were applied twice. Fungicides were applied as dilute mixtures of Captan 50WP (0.3 oz/gal) and Nova 40W (0.02 oz/gal) starting when shoot growth was approximately 5-10 in. and usually stopped during the last week in July. Fermentable sugar, the starting point for ethyl alcohol production, was estimated by two methods. One used a refractometer to measure percent soluble solids (equivalent to degrees Brix) of fresh juice obtained by pressing a repre- sentative sample of bunches through cheesecloth, and the second measured bouyant density of the juice with a Brix hydrometer. Brix readings reported are an average of read- ings obtained by the two methods. Total acidity was deter- mined by measuring the quantity of NaOH required to neu- turalize 0.17 oz (5 ml ) samples of juice prior to the start of fermentation. A minimum of two titrations for acidity were made on each juice sample. Harvest date was determined by field measurement of sugar content with a refractometer reading of approximately c 30 120 140 160 180 200 220 240 260 Julian Date Figure 1. Cumulative rainfall at Lockwood Farm, Hamden, CT, from Julian Date 126 (10 June) to 250 (7 September), 1990-1993. 17%. In some harvest years grape cultivars may have been harvested earlier than optimum in order to avoid excessive loss to birds and racoons. Although I initially collected yield and quality data on most cultivars, it became apparent that some cultivars yielded so poorly due to low vigor, poor fruit set, or bunch rots that further data were not collected. Yields were taken and calculated on a per vine basis. RESULTS AND DISCUSSION Table 1 lists cultivar, type, year planted, and juice color. By 1 993 only two vines each of Chardonnay and Riesling of the vinifera cultivars remained alive. Table 2 summarizes the evaluation of vine vigor during the first 2 years of this study. Of the 18 cultivars, Seyval, Chambourcin, Villard Blanc, Seibel 10868, and JS26-627 were the most vigorous in 1990 and 1991. Riesling was the only vinifera cultivar that exhib- ited high vigor in 1991. Complete removal of fruit clusters in 1990 allowed Vignoles, Cayuga White, Chancellor, and DeChaunac to return to high vigor in 1991. Seyval, Cham- bourcin, and Villard Blanc had high fruit set in 1990 and 1991. Seibel 10868, Horizon, and Chancellor had high fruit set in 1991. Although the four vinifera cultivars survive and grow elsewhere in Connecticut, they apparently need a more protected site than our vineyard in Hamden. Because fungi- cides were applied sparingly during 1982-1989 in order to evaluate cultivars for disease resistance, it is not clear whether disease pressure, cold weather, or a combination of both caused the low vine vigor observed in 1990 and 1991. It is possible that Cayuga White and Vidal, which are commer- cially grown in Connecticut, were harmed by problems caused by disease pressures as they showed excellent yield potential in some years but not in others (Table 3). The highest yields were observed in 1990 (Table 3), which was probably influenced by the abundant, well-timed rainfall (Fig. 1), and the thinning of bunches to two per shoot; whereas, in other years bunches were thinned to less than two per shoot. Lowest yields were observed in 1993 which can be attributed to the very low rainfall (Fig. 1) during the period berries were expanding, the great length of time between rainfall, and the fact that bunches were thinned to one per shoot. Although thinning bunches to a certain number per shoot can bias the comparative yields of culti- vars, it is likely that the highest yielding cultivars in the trial suffered the greatest yield loss by this procedure. For some cultivars three or fewer years of data were collected because there was too little or no sound fruit (Table 3). Yields cannot be compared between years because different degrees of bunch thinning were practiced. Within years cultivars were thinned and vines were trained similarly, so yields can be compared. However, this may obscure the ability of certain cultivars to support a greater fruit load. Cultivars with promising yield and outstanding vine vigor over the period, 1990-1993, were Seyval, Chambourcin, Wine Grape Trials 1990-1993 Table 2. Vine vigor and fruit set of wine grapes in 1990 and 1991. ^Cultivar Vine Vigor Juice Color 1989-90 1990-91 Fruit Set 1990 Hybrid and Non-Vinifera Grapes: Vine vigor: 0 = no pruning; 3 = extensive canewood pruned. Fruit Set: 0 = no berries set; 5 = complete berry set; R = Grape clusters removed. Table 3. Yield of selected wine grape cultivars at Lockwood Farm, Hamden, CT, 1990-1993. Cultivar 1990 Yield (lbs/vine) 1991 1992 1993 1991 Seyval white 3 3 5 5 Chambourcin red 3 3 5 5 Villard Blanc white 3 3 5 5 Seibel 10868 white 3 3 3 5 JS26-627 pink 3 3 1 3 Horizon white 2 3 4 5 Villard Noir red 2 3 3 4 Vignoles white 2 3 R 4 Chelois red 2 3 3 4 Cayuga White white 2 3 R 3 Chancellor red 2 3 R 5 Vidal 256 white 2 1 R 1 DeChaunac red 1 3 R 4 Verdelet white 1 1 R R Vinifera Grapes: Riesling white 1 3 R 2 Gamay Beaujolais red 0 0 0 0 Chardonnay white 0 0 0 0 Gewurztraminer white 0 0 0 0 Mean Seyval 16.7 9.7 8.4 2.2 9.3a Chambourcin 16.5 9.6 11.5 3.3 10.2a Villard Blanc 16.2 7.0 8.6 2.3 8.5a Horizon 6.6 5.9 10.2 2.9 6.4a Villard Noir 3.7 5.0 0.7 Chelois 3.4 4.3 Seibel 10868 4.0 4.6 JS 26-627 2.0 0.6 DeChaunac 5.4 Vignoles 3.9 Cayuga White 5.0 2.4 Vidal 8.3 4.6 a Values followed by the same letter are not significantly different by Duncans Multiple Range Test (P = 0.05). Connecticut Agricultural Experiment Station Bulletin 920 Seyval Chambourcin Villard Blanc Horizon CULTIVAR 1990 1991 [J 1992 § | 1993 8 1.5 Seyval Chambourcin Villard Blanc Horizon CULTIVAR 1 1 1990 II 1991 [I 1992 I"~l 1993 Figure 2. Estimated sugar content for fresh juice of Seyval, Chambourcin, Villard Blanc, and Horizon wine grapes harvested from 1990 to 1993. Figure 3. Total acid of fresh juice of Seyval, Chambourcin, Villard Blanc, and Horizon wine grapes harvested from 1990 to 1993. Villard Blanc, and Horizon (Fig. 2). These four cultivars yielded significantly more fruit than the others in the trial and did not differ significantly from one another in average yield when subjected to two-way analysis of variance (Table 3). Seyval, Vidal, and Cayuga White are examples of white wine grape cultivars which show excellent yield in commer- cial vineyards in Connecticut; however, there is a need for high quality, high yielding red cultivars. The red wine grape cultivar Chambourcin showed high yield potential in Hamden. Villard Noir may hold promise as a red wine grape cultivar, but its yield potential was significantly lower than that of Seyval, Chambourcin, and Villard Blanc (Table 3). JS26-627 and Seibel 10868 also showed promise as red wine grapes because of their excellent berry set, but the small, tight bunches resulted in low yield or excessive bunch rot. Chancellor also showed excellent berry set, but invariably the fruit rotted before harvest. This may have been avoided with a more intensive fungicide program. However, the four most promising cultivars followed throughout the study produced excellent fruit with low input of disease controls. Figures 2 and 3 summarize the sugar content and total acid of the juice of Seyval, Chambourcin, Villard Blanc, and Horizon before the start of fermentation. No statistics have been performed on these values because they are influenced by harvest date. In no year were our harvests timed to the optimum quality level of a cultivar. The data show that Seyval, Chambourcin, and Villard Blanc are similar in sugar content, but Chambourcin is consistently higher than Seyval and Villard Blanc in acid even when its sugar content is higher than the others (Figs. 2 and 3). It is possible that Chambourcin was harvested too early. Generally, levels of acid decline and sugar rises as the berries mature. In Hamden Seyval appears to ripen at a time near that of Villard Blanc and Chambourcin (Table 4). In contrast, Horizon appears to mature well before Seyval, Chambourcin, and Villard Blanc. This differs from observations in Geneva, NY. Thus, by the time I harvested Horizon, it may have been overripe and may account for its high sugar and low acid. The 1991 sugar data deviate the most. Excessive summer pruning and periods of drought may have resulted in insufficient canopy to produce photosynthate for the plant and fruit. Although this is specu- lation, I mention it to emphasize the necessity to strike a balance between opening of the canopy by pruning for cul- tural control of pests and leaving sufficient amounts of canopy to supply the plant with carbohydrates. CONCLUSION The unreplicated wine grape trial at Hamden, CT showed consistent high yield by Seyval, Chambourcin, Villard Blanc, and Horizon over 4 years of study, 1990-1993. The French Hybrid cultivars Chambourcin and Villard Blanc also showed sugar and acid characteristics which compared favorably with the white wine cultivar Seyval which has a proven record of hardiness and productivity in Connecticut. Further studies should focus on determining the optimum cultural conditions and harvest time for Chambourcin and Villard Blanc in Connecticut. Wine Grape Trials 1990-1993 Table 4. Harvest dates in experimental plots in Hamden, CT and expected dates in Geneva, NY (Fruit Testing Association Nursery, Inc., 1994). Cultivar Seyval Chambourcin Villard Blanc Horizon 1990 1991 1992 1993 Geneva, NY 2 Oct 24 Sept 23 Sept 22 Sept 22 Sept 21 Sept 20 Sept 21 Sept 22 Sept 12 Oct 21 Sept 24 Sept 23 Sept 22 Sept 13 Oct 7 Sept 16 Sept 23 Sept 20 Sept 29 Sept REFERENCES Baxevanis, J.J. 1989. The wines of New England. Nursery, Inc., P.O. Box 462, Geneva, NY 14456. Vinifera Wine Growers Journal 16:110-115. Walton, G.S. 1987. Testing table and wine grape varieties Fruit Testing Association Nursery, Inc. 1994. A catalog for hardiness and disease susceptibility. Connecticut Agric. of new and noteworthy fruits. Fruit Testing Association Expt. Sta, New Haven, Frontiers of Plant Science 40(l):6-7. en o CD )| The Connecticut Agricultural Experiment Station, founded in 1875, is the first experiment station in America. It is chartered by the General Assembly to make scientific inquiries and experiments regarding plants and their pests, insects, soil and water, and to perform analyses for State agencies. The laboratories of the Station are in New Haven and Windsor; its Lockwood Farm is in Hamden. Single copies of bulletins are available free upon request to Pub- lications; Box 1 106; New Haven, Connecticut 06504. ISSN 0097-0905 University of Connecticut Libraries 39153028932780