In a Nutshell:
- Hannah Breckbill, Roxane Mitten, Mark Quee, and Kyle Maxwell & Kate Solko trialed newly developed snap pea variety SL3123 against the old favorite Sugar Ann.
Key Findings
- Sugar Ann performed best for Quee and Maxwell & Solko.
- Mitten saw no yield differences between the varieties but had higher rates of germination from Sugar Ann.
- SL3123 performed best for Breckbill. In fact, Breckbill’s SL3123 had the highest yield at any farm.

Background
As fast-growing, early-season legumes, snap peas can play a helpful role in a horticultural rotation. Varieties differ in their phenologies, disease resistances and vulnerabilities, interactions with pollinators, and climate and soil preferences. Horticulture operations are also diverse, differing in their climate and soil types, agronomic practices, and labor availability. Variety trials are important tools for growers to make the optimal varietal choice. Hannah Breckbill, Roxane Mitten, Mark Quee, and Kyle Maxwell & Kate Solko were looking for snap pea varieties that produced well, tasted good, and were easy to manage. Quee wrote that “I feel like I haven't yet really figured out spring snap pea production. Hopefully with this trial I'll get a better sense of varietal differences and planting timing.” Solko wrote “I am looking to increase my bush pea production, and this trial will help me understand yield expectations going forward and select the most productive varieties within my systems. [I also want] a better understanding of whether trellising peas is important.” All the cooperators settled on growing two varieties to compare in this trial: Sugar Ann is an open pollinated (OP) classic—it won the All-American Selection prize in 1984—that is known for its sweet taste and short growing season [1], [2]. A bush variety, it can be grown free-standing, although it is suggested that trellising might increase yield. SL3132 is a new OP variety developed by Syngenta [3]. It was selected for stress tolerance. Both varieties are listed as dwarves, opening the possibility of growing them without trellising.Methods
Design

Measurements
Cooperators weighed the total amount of pea pods harvested from each plot over the course of two to three weeks. From reported plot dimensions and sowing densities, we calculated yield per square foot and yield per seed sown.
Quee estimated the number of seeds in each plot, then evaluated each plot’s germination via a qualitative scale. (‘No germination’ through ‘Very good germination’).
Mitten conducted a germination test before planting. She soaked 25 seeds of each variety, wrapped them in damp paper towel in a plastic bag, then counted the number of seeds that germinated. Based on the number of seeds that germinated in her test, she scaled the number of seeds she planted to ideally get a target population of 80 plants per plot. She then counted the actual number of plants in the plot.
Mitten and Maxwell & Solko also reported their quality assessments of the varieties’ peas.
Data analysis
We used Student’s t-test at a 95% confidence level to determine if there were significant differences between Sugar Ann and SL3123 at each farm. A t-test is used to determine if the measured values of two groups differ from each other significantly and takes into account variability seen in the replicates. 95% confidence means that if the trial were repeated in the same way we would expect to see the same results 95 times out of 100. We can perform this statistical analysis because the cooperators used completely randomized and replicated experimental designs (Figure A1).
Results and Discussion
Snap pea yield Sugar Ann performed best for Quee and Maxwell & Solko, while Breckbill had significantly higher yield from SL3123 (Figure 1). Mitten on the other hand, saw no differences between the two varieties. Breckbill’s SL3123 yield of 0.31 lb/ft2 were the highest of either variety at any farm (Figure 1). Although comparisons between farms cannot be analyzed statistically, the two farms that trellised their peas (Breckbill, Mitten) had higher yields than those two which did not. Breckbill averaged 0.26 lb/ft2; Mitten averaged 0.22 lb/ft2; Maxwell & Solko averaged 0.02 lb/ft2; Quee averaged 0.11 lb/ft2.

Conclusions and Next Steps

Appendix – Trial Design and Weather Conditions







