Abstract
Heirloom tomatoes have been shown to have a large price premium. However, in Hawaii, many do not have the appropriate disease tolerance packages and quality to be useful in the local market. We tested 29 tomato cultivars from three different market classes in two field trials from 2020 to 2021. Hybrid heirlooms had the greatest yields; traditional heirlooms had the highest quality. Introducing new lines to Hawaii will help farmers react to increasing disease pressure while still allowing for high marketable yields and creating an opportunity for price premiums.
Historically the tomato industry has operated within distinct market domains—fresh and processing (US Department of Agriculture 1983, 1991)—which has facilitated targeted research tailored to the specific needs of each supply chain. A major goal of 20th-century breeding was to develop cultivars that could endure postharvest handling throughout the supply chain (Klee and Tieman 2018). The tomato market was disrupted in the early 1980s, largely instigated by celebrity chefs and food critics (Joseph et al. 2017), which helped change consumer preferences, focusing them on heirloom types (Blanca et al. 2015). Recently, interest in improving flavor and fruit quality has received increased attention (Gao et al. 2019; Hightower 1972; Klee and Tieman 2018; Saliba-Colombani et al. 2001; Tieman et al. 2017; Zhao et al. 2019). In Hawaii, there has been interest in understanding the importance of the local market, in particular for tomato, because of a large proportion of the local market (Xu et al. 2015; Yang and Leung 2020). However, although there have been consistent increases in prices for organic, there have been inconsistent price effects for local (Xu et al. 2015; Yang and Leung 2020).
Demand for heirloom tomatoes has increased over time as they became the poster child for regional cuisine and as consumers sought more diversity in their produce. Although heirloom tomatoes have acquired numerous colloquial definitions, such as 1) being at least 50 years old, 2) being open pollinated, or 3) specific to a region with a traceable ancestry, there is no official botanical classification or accreditation system in place (Harland and Larrinua-Craxton 2009). The notion of what constitutes as heirloom in the fresh market has evolved, with sources arriving at an informal definition in terms of a cultivar’s traceable ancestry, age, and pollination status (Casañas et al. 2017; Male 1999; Watson 1996; Zeven 1998). Consumers have come to associate the term heirloom with colorful, atypically shaped, and flavorful tomatoes (Dwivedi et al. 2019). Because of the stark contrast between these characteristics and the typical commercial prototype, heirloom tomatoes have evolved into a specialty subclass within the fresh market (Grassbaugh et al. 1999; Jordan 2007; Joseph et al. 2017). New hybrid cultivars are often promoted as having heirloom-like fruit quality with modern production traits. Without established standards and regulations, the heirloom-like insignia has evolved into a phenotypic descriptor rather than a designation of heritage, genetics, or pedigree.
Although the sociocultural debate on what genuinely qualifies as an heirloom tomato may continue for years, it is possible to examine the properties that distinguish this unique market class quantitatively. The economic value attributed to heirloom tomatoes requires objective quality assurance in their labeling, with prior studies of heirloom germplasm focusing on sourcing desirable traits (Alonso et al. 2011; Dwivedi et al. 2019; Gonçalves et al. 2009; Panthee et al. 2013a; Rodríguez-Burruezo et al. 2005). We evaluated the productivity of cultivars from three market classes—traditional heirloom (TH), heirloom hybrid (HH), and commercial hybrid (CH)—to identify the most promising cultivars and market class for Hawaiian markets.
Materials and methods
Hawaii field trials
THs (cultivars recognized as heirlooms), HHs (hybrids marketed as having heirloom-like characteristics or an heirloom parent, sometimes under names such as “hyloom”), and CHs (conventionally marketed hybrids) were evaluated in field trials at the University of Hawaii research stations on Oahu in 2020 and 2021 (Table 1). To reduce the severity of pests and diseases, field tests were conducted at each location in high-tunnel screenhouses. Experiments were conducted in an augmented design to maximize accessions from each market class (Federer and Raghavarao 1975; Healy et al. 2015). An overview of the germplasm screened is presented in Table 2. Although neither experimental site was certified, management practices used during each study followed USDA National Organic Program standards. The first trial was held at the Waimanalo Research Station (lat. 21°20′7.872″N, long. 157°42′53.2188″W) in a 139-m2 modified cold frame (model 1100; Conley’s, Montclair, CA, USA) from Aug 2020 to Jan 2021. Seedlings were grown in a greenhouse at the University of Hawaii Manoa (Honolulu, HI, USA) using organic management practices and were treated with a 2% fish emulsion solution (Aqua Power™ 5-1-1; JH Biotech, Ventura, CA, USA) 1 week before transplanting. The research plot was tilled in Jul 2020, and five slightly mounded beds (0.9 × 16.8 m2) were shaped and spaced at 1.5-m centers. The experimental design consisted of five blocks (plot size, 0.9 × 1.5 m2), where four plants were considered an experimental unit. Seven check varieties were completely randomized within each block, and 17 treatment accessions were distributed among all remaining plots, each replicated once. Within row spacing, plants were offset at 60 cm apart. A single application of Suståne 8-2-4 (Suståne, Cannon Falls, MN, USA) was incorporated into each plot at planting to achieve an N rate of 224 kg·ha–1. The beds were mulched with wood chips obtained locally and installed with two driplines. Plants were supported using roller hook lines attached to two overhead high-tensile wires running along the bed perimeter. Indeterminate varieties were grown to a single stem and trained using the lower and lean method, whereas determinant types were lightly pruned and supported at the primary stem. Pest management was performed as needed and followed local extension recommendations for organic systems.
Field passport for two field trials conducted in Hawaii during 2020 and 2021.
Description of germplasm screened in two Hawaii field trials conducted from 2020 to 2021.
A second field trial was conducted at Magoon Research Station (lat. 21°18′26.748″N, long. 157°48′35.208″W) from Jun to Nov 2021. A custom-built screenhouse (92 m2) was constructed onsite and outfitted with overhead cables to facilitate the lower and lean trellising method. Transplants were produced in the same location and managed according to the previous description. The research plot was tilled, and 10 slightly mounded beds (0.6 × 5.4 m2) were shaped at 1.4-m centers. Within row spacing, plants were offset at 51 cm apart. The experimental design comprised five blocks (plot size, 0.6 × 1.4 m2) and five plants per experimental unit. Six check varieties were completely randomized within each block, and eight treatment accessions replicated once were distributed among all remaining plots. A total fertilization rate of 224 kg·ha–1 N was provided through a split application of 112 kg·ha–1 Suståne 8-2-4 at planting, whereas the remaining equivalents were supplied through soluble fertilizer (Aqua Power™ 5-1-1; JH Biotech) treatments during flower set and early harvest. All other cultural management and site preparations for the 2020 study were repeated in 2021, with the exception that all accessions evaluated were indeterminate in growth.
Data collection
Total and marketable yield
Data for yield and marketability were gathered over 10 weeks, as established by the initial harvest of each plot. The fruit were harvested as needed to guarantee that specimens for subsequent shape and quality assessments were at their peak maturity. Qualitative assessments to determine marketability were based on the grading system outlined by Healy et al. (2015). Harvest data were recorded per plot (as kilograms per plant per plot), and marketable weights were estimated at each weighing by multiplying the average fruit weight by the number of unmarketable fruit.
Analysis of tomato fruit morphology and color
Twenty fruit of each accession were harvested at the red-ripe stage (US Department of Agriculture 1991) and analyzed using the Tomato Analyzer (ver. 4.0) for 42 fruit shape and color metrics (Supplemental Table 1). Ten fruit were prepared for longitudinal and transverse imaging (Gonzalo et al. 2009). Fruit images were captured using an Epson Perfection V39 Color Photo & Document Scanner (Epson America, Inc., Los Alamitos, CA, USA). Fruit were harvested for a 10-week period. Harvest time points were established on a per-plot basis, starting on the day of the first harvest. Frequency was as needed, because only red-ripe fruit were used for images and chemical analysis.
Evaluation of fruit physicochemical properties
Fruit prepared for image analysis were examined for the following fruit quality attributes: total titratable acidity (TTA), total soluble solids (reported as a percentage of degrees Brix), and pH. Cross sections of approximately nine fruit were homogenized in a blender, and three 50-mL aliquots were stored at –20 °C (Supplemental Fig. 1). Before analysis, samples were thawed for 4 h in the refrigerator (4 °C) and then brought to room temperature (20 to 22 °C) (Casals et al. 2019). Each aliquot was rehomogenized and strained with a cheesecloth to remove excess solids. Brix values were estimated using a digital refractometer (model 30051; Sper Scientific, Scottsdale, AZ, USA) with automatic temperature compensation and recorded as the average of three technical replicates. Using the procedures described in Panthee et al. (2013b), TTA was estimated as the percentage of citric acid (CA) by volume, and pH was measured using a mini automatic titrator and pH meter (model 84432; Hanna Instruments, Woonsocket, RI, USA).
Data analysis
All statistical analyses were conducted using the R statistical software (ver. 4.2.0; R Foundation for Statistical Computing, Vienna, Austria). Each environment’s total and marketable yield data were calculated as the average kilograms per plant per plot (Supplemental Tables 2–8). As a result of significant variations across sites, each field experiment was analyzed separately. Total and marketable yields were evaluated by fitting a linear mixed model with genotype as the fixed effect and block as the random effect. The R package augmentedRCBD (Aravind et al. 2020) was used to conduct an analysis of variance and least square mean comparisons using Fisher’s least significant difference when P < 0.05.
Results
During the 2020 and 2021 field trials in Hawaii, 29 cultivars (11 HH, 11 CH, and 6 TH) were tested for total yield (kilograms per plant), marketable yield (kilograms per plant), and quality (Fig. 1).
2020 Trial
In 2020, there were large differences in marketable yield (Tables 3 and 4), with Skyway, a commercial hybrid, having the highest yield, whereas Stealth, an heirloom hybrid, had the lowest (Table 4). The check cultivars did not perform among the top two groups in this trial, suggesting the large potential for new-cultivar introduction in Hawaii. Cultivars of all market classes had representatives that showed commercially viable yields, and there were few changes in ranking between marketable yields and total yields (Table 4). For the fruit quality traits TTA and Brix (Fig. 2), THs’ mean TTA (0.56% CA) was significantly greater than CHs (0.43% CA, P = 5.12 × 10−10) and HHs (0.44% CA, P = 6.94 × 10−8). There was less variation among class means for Brix, with values for CHs (3.93%) roughly equivalent to the mean for THs (3.94%) (Fig. 2). THs had more negatively skewed distributions, whereas the distribution for CHs was positively skewed (Supplemental Table 3). Among the market types, CH was the only group that was significantly different (P = 0.03) from HHs.
Analysis of variance for block- and treatment-adjusted total and marketable yields in the Waimanalo 2020 field trial.
Estimated marginal means for total and marketable yield for the Waimanalo 2020 field trial.
Mean comparison between Waimanalo 2020 market class values using the Games-Howell test (Holmes adjusted P < 0.05) for two fruit quality traits: (A) total titratable acidity [measured as the percentage of citric acid (CA)] and (B) degrees Brix.
Citation: HortTechnology 35, 2; 10.21273/HORTTECH05580-24
2021 Trial
In 2021, there were large differences in yield (P ≤ 0.01; Tables 5 and 6). The HH Quasimodo significantly outperformed all other varieties (Table 6). The check cultivars performed much better in this trial compared with experimental treatments (Table 6). Again, there was no clear pattern based on market class with respect to yield, with all market classes represented in good-yielding cultivars. There were also minimal differences in rank between marketable yield and total yield. In 2021, mean comparisons for TTA exhibited a comparable ranking among market classes, with THs having the highest mean (0.73% CA) and CHs having the lowest (0.65% CA) (Fig. 3). Differences among the market classes were seen between THs (0.73% CA) and CHs (0.65% CA; P = 6.30e–3), and HHs and CHs (0.71% CA; P = 0.02). In contrast, CHs had the highest mean Brix (4.34%), and no significant differences were identified among the market classes. Although the two trials were not compared statistically, trends were observed for the varieties replicated across environments. The HH Stealth had the lowest total yield. The TH Old German was among the highest yielding lines and it outperformed the two commercial checks included in the trial.
Analysis of variance for block- and treatment-adjusted total and marketable yields in the Honolulu 2021 field trial.
Estimated marginal means for total and marketable yield for the Honolulu 2021 field trial.
Mean comparison between Honolulu 2021 market class values using the Games-Howell test (Holmes adjusted P < 0.05) for two fruit quality traits: (A) total titratable acidity [measured as the percentage of citric acid (CA)] and (B) degrees Brix.
Citation: HortTechnology 35, 2; 10.21273/HORTTECH05580-24
Discussion
Total and marketable yields were different among the three market classes (TH, HH, and CH). CHs were among the top-performing lines in both yield metrics in 2020 (Waimanalo), with yields being greatest from this market class found in the determinate types (Skyway, Cypress, and Grebe), suggesting that trellising or pruning techniques may have contributed to the results. We did not restrict the genotypes included to those with disease resistance qualities generally required by local producers (Tomato yellow leaf curl virus and Tomato spotted wilt virus). Consequently, an unexpected finding was the strong performance of the susceptible HH variety Quasimodo, which was the top-performing line in 2021 (Honolulu). Furthermore, its total and marketable yields in the 2020 trial were not statistically different from those of the top two CH cultivars. Ginfizz was susceptible to disease yet produced consistent and early yields.
Environmental conditions were most likely a significant factor in this study; therefore, additional field studies will be required to make sound recommendations to local growers. The traditional heirloom Old German provided the most unexpected findings in both field studies, as its total and marketable yields surpassed several CHs. Although the scale and cultural practices used in the field trials may not be representative of more intensive commercial operations, these results suggest that a comparative economic analysis of specialty market types with varying disease resistance may be a fruitful area of research for cooperative extension in Hawaii.
Heirloom tomatoes have come to personify the preferred fresh-market archetype and command a substantial premium in the marketplace (Healy et al. 2015; Navazio 2012). Many distinctive traits of heirloom fruit are visible cues typically marketed to consumers; however, customers cannot discern between products based on the quality of their internal components. There have been consistent increases in tomato prices as a result of being designated an heirloom (Dwivedi et al. 2019). Therefore, if a tomato has the exterior traits of an heirloom, it will likely be viewed as such, but it may lack the intrinsic quality that customers anticipate. This presents a significant ethical concern regarding heirloom labeling, especially given the consumer demand for heirloom tomatoes (Alexander 2006; Klee & Tieman, 2018; Sydorovych et al. 2013). Given that red tomatoes remain the commercial standard in most markets, the differences in fruit color reflected in heirloom varieties have come to typify the label (Dwivedi et al. 2019; Joseph et al. 2017; Rodríguez-Burruezo et al. 2005). Although postharvest handling and shelf life were not investigated explicitly, they are crucial factors that affect fruit quality and have also influenced historical perceptions of heirloom tomatoes.
Conclusion
There were clear differences among CHs, HHs, and THs as classes; however, individual genotypes within each class performed well, thus providing many options for growers in Hawaii. Although measuring the chemical composition and perception of flavor was outside the scope of our study, it represents an intriguing area for future research and remains an essential element required to define the heirloom archetype. The implications of this field of study may provide a spectrum of desirable traits that can be targeted in future breeding programs, can broaden the definitions of what makes a cultivar heirloom, and can enhance ongoing efforts to improve fresh-market tomato quality.
References cited
Alexander W. 2006. The $64 tomato: How one man nearly lost his sanity, spent a fortune, and endured an existential crisis in the quest for the perfect garden. Algonquin Books, Chapel Hill, NC, USA.
Alonso A, Salazar JA, Arroyo A, Grau A, García-Martínez S, Serrano M, Ruiz JJ. 2011. Screening a diverse collection of heirloom tomato cultivars for quality and functional attributes. Acta Hortic. 918:551–555. https://doi.org/10.17660/ActaHortic.2011.918.69.
Aravind J, Mukesh Sankar S, Wankhede DP, Kaur V. 2020. Augmented RCBD: Analysis of augmented randomised complete block designs. R package version 0.1, 2.
Blanca J, Montero-Pau J, Sauvage C, Bauchet G, Illa E, Díez MJ, Francis D, Causse M, van der Knaap E, Cañizares J. 2015. Genomic variation in tomato, from wild ancestors to contemporary breeding accessions. BMC Genomics. 16:257.
Casals J, Rull A, Segarra J, Schober P, Simó J. 2019. Participatory plant breeding and the evolution of landraces: a case study in the organic farms of the Collserola natural park. Agronomy. 9(9):486.
Casañas F, Simó J, Casals J, Prohens J. 2017. Toward an evolved concept of landrace. Front Plant Sci. 8:145. https://doi.org/10.3389/fpls.2017.00145.
Dwivedi S, Goldman I, Ortiz R. 2019. Pursuing the potential of heirloom cultivars to improve adaptation, nutritional, and culinary features of food crops. Agronomy. 9(8):441. https://doi.org/10.3390/agronomy9080441.
Federer AWT, Raghavarao D. 1975. On augmented designs. Biometrics. 31(1):29–35. https://doi.org/10.2307/2529707.
Gao L, Gonda I, Sun H, Ma Q, Bao K, Tieman DM, Burzynski-Chang EA, Fish TL, Stromberg KA, Sacks GL, Thannhauser TW, Foolad MR, Diez MJ, Blanca J, Canizares J, Xu Y, van der Knaap E, Huang S, Klee HJ, Giovannoni JJ, Fei Z. 2019. The tomato pan-genome uncovers new genes and a rare allele regulating fruit flavor. Nat Genet. 51(6):1044–1051. https://doi.org/10.1038/s41588-019-0410-2.
Gonçalves LSA, Rodrigues R, do Amaral Júnior AT, Karasawa M, Sudré CP. 2009. Heirloom tomato gene bank: Assessing genetic divergence based on morphological, agronomic and molecular data using a Ward-modified location model. Genet Mol Res. 8(1):364–374. https://doi.org/10.4238/vol8-1gmr549.
Gonzalo MJ, Brewer MT, Anderson C, Sullivan D, Gray S, van der Knaap E. 2009. Tomato fruit shape analysis using morphometric and morphology attributes implemented in tomato analyzer software program. J Am Soc Hortic Sci. 134(1):77–87. https://doi.org/10.21273/JASHS.134.1.77.
Grassbaugh E, Harker T, Bergefurd B, Bennett M. 1999. Heirloom tomato production and marketing 1995–1998. Hortic Crop Sci. 684:14.
Harland G, Larrinua-Craxton S. 2009. The tomato book: A guide to the pleasures of choosing, growing, and cooking (1st ed). Dorling Kindersley, London.
Healy K, Theisen T, Dawson J. 2015. Tomato variety trials for direct market quality and flavor. Presented at the Proceedings of the Organic Agriculture Research Symposium, LaCrosse, WI, USA, 25–26 Feb 2015.
Hightower J. 1972. Hard tomatoes, hard times: Failure of the land grant college complex. Society. 10:10–22. https://doi.org/10.1007/BF02695245.
Joseph H, Nink E, McCarthy A, Messer E, Cash SB. 2017. The heirloom tomato is ‘in’: Does it matter how it tastes? Food Culture Soc. 20(2):257–280. https://doi.org/10.1080/15528014.2017.1305828.
Jordan JA. 2007. The heirloom tomato as a cultural object: Investigating taste and space. Sociol Ruralis. 47(1):20–41. https://doi.org/10.1111/j.1467-9523.2007.00424.x.
Klee H, Tieman D. 2018. The genetics of fruit flavour preferences. Nat Rev Genet. 19(6):347–356. https://doi.org/10.1038/s41576-018-0002-5.
Male CJ. 1999. One hundred heirloom tomatoes for the American garden. Workman Publishing, New York, NY, USA.
Navazio J. 2012. The organic seed grower: A farmer’s guide to vegetable seed production. Chelsea Green Publishing, White River Junction, VT, USA.
Panthee DR, Labate JA, McGrath MT, Breksa AP, Robertson LD. 2013a. Genotype and environmental interaction for fruit quality traits in vintage tomato varieties. Euphytica. 193(2):169–182. https://doi.org/10.1007/s10681-013-0895-1.
Panthee DR, Labate JA, Robertson LD. 2013b. Evaluation of tomato accessions for flavour and flavour-contributing components. Plant Genet Resour. 11(2):106–113. https://doi.org/10.1017/S1479262112000421.
Rodríguez-Burruezo A, Prohens J, Roselló S, Nuez F. 2005. Heirloom varieties as sources of variation for the improvement of fruit quality in greenhouse-grown tomatoes. J Hortic Sci Biotechnol. 80(4):453–460. https://doi.org/10.1080/14620316.2005.11511959.
Saliba-Colombani V, Causse M, Langlois D, Philouze J, Buret M. 2001. Genetic analysis of organoleptic quality in fresh market tomato: 1. Mapping QTLs for physical and chemical traits. Theor Appl Genet. 102(2–3):259–272. https://doi.org/10.1007/s001220051643.
Sydorovych O, Rivard CL, O’Connell S, Harlow CD, Peet MM, Louws FJ. 2013. Growing organic heirloom tomatoes in the field and high tunnels in North Carolina: Comparative economic analysis. HortTechnology. 23(2):227–236. https://doi.org/10.21273/HORTTECH.23.2.227.
Tieman D, Zhu G, Resende MFR, Lin T, Nguyen C, Bies D, Rambla JL, Beltran KSO, Taylor M, Zhang B, Ikeda H, Liu Z, Fisher J, Zemach I, Monforte A, Zamir D, Granell A, Kirst M, Huang S, Klee H. 2017. A chemical genetic roadmap to improved tomato flavor. Science. 355(6323):391–394. https://doi.org/10.1126/science.aal1556.
USDA. 1983. United States standards for grades of tomatoes for processing. https://doi.org/10.1533/9781845696122.appendix2. [accessed 14 Feb 2025].
USDA. 1991. United States standards for grades of fresh tomatoes. https://www.ams.usda.gov/sites/default/files/media/Tomato_Standard%5B1%5D.pdf. [accessed 14 Feb 2025].
Watson B. 1996. Taylor’s guide to heirloom vegetables. Houghton Mifflin Harcourt. Boston, MA, USA.
Xu X, Keahiolalo K, Loke MK, Leung P. 2015. Local premium or local discount: The case of packaged fresh tomatoes in Hawaii. J Agric Appl Econ. 47(3):345–358. https://doi.org/10.1017/aae.2015.11.
Yang Y, Leung P. 2020. Price premium or price discount for locally produced food products? A temporal analysis for Hawaii. J Asia Pacific Economy. 25(4):591–610. https://doi.org/10.1080/13547860.2019.1687250.
Zeven ACC. 1998. Landraces: A review of definitions and classifications. Euphytica. 104(2):127–139. https://doi.org/10.1023/A:1018683119237.
Zhao J, Huang S, Bitton F, Liu D, Causse M, Zhao J, Bauchet G, Tieman D, Sauvage C, Klee H. 2019. Meta-analysis of genome-wide association studies provides insights into genetic control of tomato flavor. Nat Commun. 10:1534. https://doi.org/10.1038/s41467-019-09462-w.