Research Note Yield Components and Fruit Composition of Five Barbera Grapevine Selections in the San Joaquin Valley, California

Matthew W. Fidelibus,1* L. Peter Christensen,1 Deborah A. Golino,2 Nancy L. Sweet,2 and Kimberley A. Cathline1

Abstract: Barbera grapevine (Vitis vinifera) selections from Foundation Plant Services (FPS), University of California, Davis, were evaluated in Parlier, Fresno County, California. Green potted plants of Barbera, FPS selections 02, 03, 04, 05, and 06, were planted in the summer of 2000, and fruit yield and quality were assessed annually between 2003 and 2006. Selection 02 performed similarly to several other selections with respect to fruit composition and yield, but its berries were 10 to 25% heavier than the others. Large berries made the clusters of FPS 02 excessively compact; the incidence of sour rot was sometimes two to three times higher for FPS 02 than for other selections. Selections 03 and 05 originated from the same mother vine and performed identically. Their fruit composition was similar to several other selections, but they had the highest, or among the highest, yield of the selections tested and were moderately susceptible to sour rot. Selection 04 was generally similar to selections 03 and 05 except that 04 had lower yields in two of the four years tested. Selection 06 was distinctive in that it was generally the earliest ripening, had the smallest berries, the lowest yields, the lowest yield:pruning weight ratios, and was the least susceptible to sour rot. Key words: clones, fruit quality, germplasm, grapes, Vitis vinifera L.

Barbera is the second most widely planted red winegrape programs in Europe where evaluation is supported with in Italy, and its tendency to produce good yields of fruit public funds. with relatively high acidity have helped establish Barbera The first indexed selection of Barbera in California, as an important cultivar in several other countries, includ- Barbara FPS 01, was added to the list of registered variet- ing the United States (Christensen, 2003, Ferrandino et al. ies in 1970 and constituted much of the plantings during 2007, Mannini 1995). Barbera has been grown in Califor- the following decades (Christensen 2003). In 1982, Barbera nia since the 1880s, and extensive plantings were made in FPS 02 was introduced from Rauscedo, Italy. Because there the San Joaquin Valley during the 1970s and 1980s (Chris- was no experience with this selection, a trial comparing tensen 2003). Presently there are ~2,873 ha of Barbera in selections 01 and 02, the only indexed selections at the California, 90% of which is located in the San Joaquin Val- time, was established in Fresno County (L.P. Christensen, ley (CDFA 2009). unpublished data, 1995). Selection 02 was found to be par- During Barbera’s long history of cultivation, many clones ticularly productive, but its exceptionally large berries and have been selected but few have been systematically evalu- clusters contributed to later ripening and more sour rot in ated (Mannini 1995). In the United States, the majority of comparison to selection 01 (L.P. Christensen, unpublished registered grape clones are released as selections by Foun- data, 1995; Christensen 2003). Unfortunately, after bud- dation Plant Services (FPS), College of Agricultural and wood of Barbera FPS 01 was found to have leafroll virus in Environmental Sciences, University of California, Davis. 1983, that selection was removed from the list of registered The FPS normally screens new selections for target virus vines, leaving FPS 02 as the only registered Barbera selec- diseases but does not evaluate selections, unlike similar tion in the California Grapevine Registration & Certifica- tion Program (Christensen 2003). The FPS Barbera collection was augmented between 1Department of Viticulture and Enology, University of California, Davis, CA 1993 and 2001. In 1993, FPS received several selections 95616, and Kearney Agricultural Center, Parlier, CA 93648, and 2Foundation from Torino, Italy, that were released in 2001 as FPS 03, Plant Services, University of California, Davis, CA 95616. 04, and 05. Selections 03 and 05 were collected from the *Corresponding author (email: [email protected]) same vine but were considered separate selections, in ac- Acknowledgments: This research was made possible by grants from the cordance with the FPS policy at that time. In 1996, a source American Vineyard Foundation and the Viticulture Consortium West. of FPS 01 that tested free of virus was identified and rein- The assistance of Donald Katayama, Jorge Osorio-Aguilar, Pierre-Thibaut troduced in 2001 as FPS 06. Based on previous work (L.P. Verdenal, and Karin Kawagoe is gratefully acknowledged. Christensen, unpublished data, 1995), it was anticipated Manuscript submitted May 2009, revised Jun 2009, accepted Jul 2009. Publica- tion costs of this article defrayed in part by page fees. that selection 06 would be superior to selection 02, but Copyright© 2009 by the American Society for Enology and Viticulture. All there were no data to determine the relative merits of FPS rights reserved. 03, 04, and 05. Given that selections 03 and 05 represented 533 Am. J. Enol. Vitic. 60:4 (2009) 534 – Fidelibus et al. cuttings from different canes from the same source vine, it an automatic titrator (900A, Orion Research). Clusters on was assumed that the two selections should exhibit similar each vine were then harvested. Incidence of sour rot was characteristics. Confirmation from a replicated trial could recorded, clusters were weighed, and number of berries per provide the justification needed to retire one of those selec- cluster were calculated. Sour rot incidence was evaluated tions. Thus, the objective of this study was to compare the based on the proportion of clusters having four or more performance of the five FPS Barbera selections in Fresno adjoining berries with decay (Fidelibus et al. 2005). County, California. Data were subjected to analysis of variance using the GLM procedure of SAS software (SAS Institute, Cary, NC). Materials and Methods Block, year, block by year interaction, and block by treatment Own-rooted plants of registered Barbera selections FPS interaction were considered random effects. When treatment 02, 03, 04, 05, and 06 were obtained from FPS. Selections effects were significant (p ≤ 0.05), treatment means were have the following origins: FPS 02, Rauscedo 6, Italy; FPS separated by Duncan’s new multiple range test. 03, CVT 171, Torino, Italy; FPS 04, CVT 84, Torino, Italy; FPS 05, CVT 171, Torino, Italy; and FPS 06 Marshall (32 Results and Discussion V7), a selection of FPS 01. The vines were planted in 2000 In 2006, the only year in which phenological data were to a vineyard at the Kearney Agricultural Center. Selections collected, all selections began growth on about 21 Mar and were planted in a randomized, complete block design with bloomed on about 14 May. Veraison occurred on 10 July for one vine of each selection in each of 10 blocks. The vines every selection except 06, which began veraison on 15 July. were spaced ~2.3 m within rows, and 3 m between rows. Over the course of the study, the fruit of selection 06 had The vineyard soil was classified as a Hanford fine sandy similar or greater soluble solids at harvest than the fruit of loam with ~1.5 m rooting depth (USDA 1971). Each year, other clones (Table 1). This finding was true even in 2006 all vines were subjected to cultural practices that were con- when selection 06 was the last to begin ripening but its sidered typical for winegrapes in the San Joaquin Valley. fruit was harvested one week earlier than the fruits of the The second winter after planting, each grapevine was other selections. The other selections had similar soluble pruned to two canes that were retained as permanent bi- solids except in 2005, when selection 03 had less soluble lateral cordons, supported by a single wire mounted on a solids than selection 04, and in 2006, when selection 04 had trellis at 1.35 m above the vineyard floor. The trellises also more soluble solids than selections 02 or 05. had one foliar catch wire ~0.3 m above the cordon wire. Juice pH and titratable acidity differed among the selec- Shoot thinning was performed the following spring, leav- tions, as would be expected from the soluble solids data ing 8 to 9 shoots per cordon for permanent spur positions. (Table 2). Selection 06 consistently had similar or higher Each cordon was ~1 m long, with a 0.3 m gap between the pH than the other selections. FPS 03, 04, and 05 had simi- ends of adjacent cordons to facilitate the identification and lar juice pH levels, except in 2005 when selection 03 had harvest of the single-vine plots. Each subsequent winter, lower pH than selections 04 and 05 and in 2006 when se- vines were pruned to 16–18 two-node spurs per vine. In lections 04 and 06 had higher juice pH than the other se- spring, shoot thinning was conducted to leave two shoots lections. Similarly, juices from selection 06 generally had per spur. Flower clusters were counted after shoot thinning, similar or lower titratable acidity than juices from other in spring 2003, and every year thereafter through 2006, the selections, whereas juices from the other selections gener- last season data were collected. ally had similar titratable acidity (Table 2). Collectively, In 2006, each vine was inspected daily around budbreak, the fruit composition data agreed with previous findings bloom, and veraison. When 50% of the organs on a vine (L.P. Christensen, unpublished data, 1995), in which the reached each of those stages, the day of year was recorded fruit of FPS 01 generally matured before that of FPS02. In for statistical analysis. In the winter of 2005 and 2006, the prunings of each vine were weighed and yield:pruning Table 1 Soluble solids of grape juices from Barbera grapevine weight ratios were calculated. selections, Parlier, CA, 2003–2006. Vines of all selections were harvested on 4 Sept 2003, Soluble solids (Brix) 31 Aug 2004, and 26 Sept 2005. In 2006, vines of all other Selection 2003 2004 2005 2006 4-yr avg. selections were harvested on 13 Sept, except for selection 02 24.9 22.3 ba 23.3 bc 23.4 b 23.4 b 06, which was harvested earlier, on 6 Sept, in an attempt to 03 25.3 21.9 b 22.2 c 24.6 ab 23.5 b match soluble solids. On each harvest date, 100 berries per 04 24.9 22.2 b 23.6 b 25.0 a 23.9 ab vine were randomly collected from throughout the canopy 05 25.0 22.3 b 23.3 bc 23.4 b 23.5 b of each vine. The samples were then weighed and homog- 06 24.4 24.2 a 24.7 a 23.8 ab 24.3 a enized in a blender. Soluble solids from their filtered juice Signif were measured with a temperature-compensating digital re- Year (Y) <0.001 fractometer (Palette 101, Atago, Farmingdale, NY). Juice Selection (S) 0.303 0.004 0.002 0.044 0.020 pH was measured with a meter (SA720, Orion Research, Y • S <0.001 Boston, MA) and titratable acidity (TA) was determined aValues within a column followed by a different letter are significantly by titration with 0.133 N NaOH to an 8.20 end point using different according to Duncan’s new multiple range test.

Am. J. Enol. Vitic. 60:4 (2009) Barbera Selections – 535

Table 2 The pH and titratable acidity of grape juices from Barbera grapevine selections, Parlier, CA, 2003–2006. pH Titratable acidity (g/100 mL) Selection 2003 2004 2005 2006 4-yr avg. 2003 2004 2005 2006 4-yr avg. 02 3.55 3.26 ba 3.52 bc 3.40 b 3.43 c 0.690 a 0.752 a 0.929 ab 1.007 a 0.845 a 03 3.56 3.28 b 3.46 c 3.45 b 3.44 bc 0.665 ab 0.743 a 0.954 a 0.951 ab 0.828 a 04 3.56 3.28 b 3.54 b 3.53 a 3.48 b 0.644 b 0.720 a 0.860 bc 0.879 b 0.778 b 05 3.55 3.20 b 3.55 b 3.44 b 3.43 c 0.663 ab 0.733 a 0.893 ab 0.978 a 0.817 a 06 3.58 3.46 a 3.64 a 3.51 a 3.55 a 0.680 ab 0.612 b 0.798 c 0.887 b 0.744 c Signif Year (Y) <0.001 <0.001 Selection (S) 0.679 0.001 <0.001 <0.001 <0.001 0.137 <0.001 <0.001 0.006 <0.001 Y • S <0.001 0.008 aValues within a column followed by a different letter are significantly different according to Duncan’s new multiple range test. the present study, FPS 06, derived from FPS 01, showed those selections having more berries per cluster (Table 5). consistently earlier ripening than FPS 02. Selection 02 always had the heaviest berries, whereas ber- Difference in yield among selections varied year to year. ries of selections 03, 04, and 05 were less heavy and those However, in every year, selections 03 and 05 produced 20 of selection 06 were the lightest (Table 5). to 30% more fruit, by weight, than selection 06. In two of Barbera’s large berries can cause its clusters to become four years, 03 and 05 produced 20% more fruit than selec- excessively compact and therefore susceptible to rot (Man- tion 04 (Table 3). The higher yields of selections 03 and 05 nini 1995). Selection 02, which consistently had the larg- are attributable to the production of more clusters (2003 est berries, was the most susceptible to sour rot (Table 6). and 2006) or heavier clusters (2004 through 2006) than In contrast, selection 06, which often had the fewest and some of the other selections (Table 4). The greater cluster smallest berries, usually had the lowest rot incidence. Sour weights of selections 03 and 05 were most likely due to rot incidence was very low in 2004, perhaps partly because the fruit were harvested earlier (at lower soluble solids) than in any other year. Susceptibility of grapes to the second- Table 3 Yield of Barbera selections, Parlier, CA, 2003–2006. ary infections that comprise the sour rot complex increases Yield per vine (kg) with fruit maturity (Marois et al. 1992). Selection 2003 2004 2005 2006 4-yr avg. Selection 06 had the heaviest pruning weights, which 02 11.6 ba 25.2 b 27.2 b 22.9 a 21.7 b were 30 to 40% greater than those of the other selections 03 15.1 a 29.7 a 33.1 a 23.6 a 25.4 a (Table 7). Because of its comparatively low yields and high 04 10.2 b 24.2 bc 31.7 a 19.3 b 21.5 b pruning weights, selection 06 had the lowest yield:pruning 05 13.7 a 27.0 ab 32.5 a 24.5 a 24.4 a weight ratio of the selections tested. Differences among 06 11.7 b 20.8 c 25.0 b 16.8 b 18.6 c the other selections with respect to yield:pruning weight Signif ratio were inconsistent, but all selections, including 06, Year (Y) <0.001 had yield:pruning weight ratios that suggested moderate to Selection (S) <0.001 <0.001 <0.001 <0.001 <0.001 moderate-to-high vigor (Smart and Robinson 2001). Grow- Y • S 0.025 ers who plant selection 06 might consider leaving higher aValues within a column followed by a different letter are significantly node numbers at pruning to compensate for the higher shoot different according to Duncan’s new multiple range test. and cane mass and the lighter crop load.

Table 4 Clusters per vine and cluster weight of Barbera grapevine selections, Parlier, CA, 2003–2006. Clusters per vine (no.) Cluster wt (kg) Selection 2003 2004 2005 2006 4-yr avg. 2003 2004 2005 2006 4-yr avg. 02 40 bca 77 114 61 ab 73 c 0.28 0.32 ab 0.24 b 0.38 a 0.31 ab 03 49 a 88 117 65 a 80 a 0.31 0.34 a 0.29 a 0.36 a 0.32 a 04 38 c 82 119 57 b 74 bc 0.27 0.30 bc 0.27 a 0.34 a 0.29 b 05 47 a 87 113 68 a 79 ab 0.29 0.31 ab 0.29 a 0.36 a 0.31 a 06 44 ab 78 113 61 ab 74 bc 0.27 0.27 c 0.22 b 0.28 b 0.26 c Signif Year (Y) <0.001 <0.001 Selection (S) <0.001 0.092 0.894 0.032 0.010 0.059 0.004 <0.001 <0.001 <0.001 Y • S 0.522 0.032 aValues within a column followed by a different letter are significantly different according to Duncan’s new multiple range test.

Am. J. Enol. Vitic. 60:4 (2009) 536 – Fidelibus et al.

Table 5 Berries per cluster and berry weight of Barbera grapevine selections, Parlier, CA, 2003–2006. Berries/cluster (no.) Berry wt (g) Selection 2003 2004 2005 2006 4-yr avg. 2003 2004 2005 2006 4-yr avg. 02 120 ba 116 bc 62 c 121 ab 105 c 2.39 a 2.82 a 3.87 a 3.15 a 3.06 a 03 142 a 140 a 94 a 131 a 127 a 2.17 b 2.40 b 3.06 bc 2.75 b 2.59 b 04 123 b 126 ab 92 a 122 ab 115 b 2.23 b 2.35 b 2.92 d 2.77 b 2.57 b 05 135 ab 128 ab 93 a 131 a 122 ab 2.16 b 2.45 b 3.10 b 2.75 b 2.61 b 06 135 ab 109 c 75 b 112 b 108 c 1.98 c 2.46 b 2.96 cd 2.48 c 2.47 c Signif Year (Y) <0.001 <0.001 Selection (S) 0.030 <0.001 <0.001 0.039 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Y • S 0.012 <0.001 aValues within a column followed by a different letter are significantly different according to Duncan’s new multiple range test.

Table 6 Sour rot incidence among clusters of grapes from Table 7 Pruning weights and yield: pruning weight ratio different Barbera grapevine selections, Parlier, CA, 2003–2006. from Barbera grapevine selections, Parlier, CA, 2005–2006. Sour rot incidence (%) Pruning weight Yield: pruning weight Selection 2003 2004 2005 2006 4-yr avg. (kg/vine) (kg/kg) 02 28 aa 4 60 a 69 a 40 a 2-yr 2-yr Selection 2005 2006 avg. 2005 2006 avg. 03 13 b 3 49 b 50 b 29 b 02 4.8 ba 3.6 b 4.21 b 5.8 b 6.8 a 6.30 bc 04 15 b 1 47 b 35 c 25 b 03 4.6 b 3.6 b 4.12 b 7.6 a 7.1 a 7.40 a 05 12 b 3 47 b 43 bc 26 b 04 5.2 b 4.0 b 4.56 b 6.5 b 5.1 b 5.79 c 06 22 ab 0 32 c 21 d 19 c 05 5.2 b 3.6 b 4.42 b 6.6 b 7.3 a 6.96 ab Signifi 06 8.0 a 5.3 a 6.66 a 3.5 c 3.6 c 3.55 d Year (Y) <0.001 Signifi Selection (S) 0.036 0.054 <0.001 <0.001 <0.001 Year (Y) <0.001 0.895 Y • S <0.001 Selection (S) <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Y • S 0.030 0.018 aValues within a column followed by a different letter are significantly a different according to Duncan’s new multiple range test. Values within a column followed by a different letter are significantly different according to Duncan’s new multiple range test.

Conclusions Ferrandino, A., S. Guidoni, and F. Mannini. 2007. Grape quality Barbera FPS 06 produced lower yields than many of the parameters and polyphenolic content of different ‘Barbera’ and other selections. However, selection 06 may nevertheless ‘Nebbiolo’ (Vitis vinifera L.) clones as influenced by environmental conditions—preliminary results. Acta Hortic. 754:437-442. be the most desirable selection tested for growers in warm climates because its berries are smaller, its clusters are less Fidelibus, M.W., L.P. Christensen, D.G. Katayama, and P.T. Verdenal. 2005. Performance of Zinfandel and Primitivo grapevine selections susceptible to sour rot, and its fruit composition remains in the Central San Joaquin Valley, California. Am. J. Enol. Vitic. comparable to most selections. FPS 03 and 05 were almost 56:284-286. identical with respect to every variable measured, regard- Mannini, F. 1995. Grapevine clonal selection in Piedmont (North- less of year, as would be expected given that they were west Italy): Focus on Nebbiolo and Barbera. In Proceedings of the collected from the same vine. These data suggest that one International Symposium on Clonal Selection. J.M. Rantz (ed.), pp. or the other selection could be withdrawn from the FPS 20-32. Am. Society for Enology and Viticulture, Davis. catalog. Selection 02 was the least desirable selection be- Marois, J.J., A.M. Bledsoe, and L.J. Bettiga. 1992. Miscellaneous cause of large berries and high rot potential. secondary invaders and sour rot. In Grape Pest Management. D.L. Flaherty et al. (eds.), p. 69. University of California, Agricultural Literature Cited and Natural Resources, Oakland. Smart, R., and M. Robinson. 2001. Sunlight into Wine. Winetitles, CDFA. 2009. Grape Acreage Report: 2008 Crop. USDA-NASS, Cali- Adelaide. fornia Field Office, Sacramento. USDA. 1971. Soil Survey, Eastern Fresno Area, California. USDA Christensen, L.P. 2003. Barbera. In Wine Grape Varieties in California. Soil Conservation Service. US Government Printing Office, Wash- L.P. Christensen et al. (eds.), pp. 25-27. University of California, ington, DC. Agricultural and Natural Resources, Oakland.

Am. J. Enol. Vitic. 60:4 (2009)