Jake Sanne*, Dillon P. Golding**, Peter Arnold*, Jenna Beville*, Derek Hilfiker*, Forrest Brown*, and Jonathan O. C. Kubesch*, ***
*Virginia Tech School of Plant and Environmental Sciences
**Hoot Owl Hollow Farm, Woodlawn, VA
***Country Home Farms, Pembroke, VA
Introduction
Buffalo clover (Trifolium reflexum), is a rare native clover present in the eastern United States. In Kentucky, this species occurs in the vicinity of Mammoth Cave, as well as further west. Conservation efforts in the eastern United States have maintained many of these native populations, though there is interest in using horticulture and agriculture to increase buffalo clover populations (Quesenberry et al., 2003; Kubesch, 2020; Kubesch et al., 2022).
A limitation of these alternative approaches is the lack of data regarding the establishment of native clovers. For many rare plants, plugs are grown in nurseries and then planted to field sites (Littlefield, 2022). Even in the only published horticultural research, seed was grown into plugs for field plantings (Quesenberry et al., 2003). Buffalo clover has a fair degree of phenotypic variability in growth form, life history, and flower color across the geographic range. These differences in plant material might also suggest diversity in seed characteristics.
Seed weight is an important seed characteristic for increasing buffalo clover populations. Seed weights are associated with increased establishment success (Catano et al., 2022; Westoby, 1998). Published seed weights of red (Trifolium pratense), white (Trifolium repens), and buffalo clover come from a publication using buffalo clover accessions from the Coastal Plain (Quesenberry et al., 2003). Coastal Plain accessions appear to have annual life histories whereas Ohio River Valley accessions appear to be short-lived biennials or perennials (Kubesch, unpublished observations). The present study sought to map and determine seed weights for buffalo clover. The authors hypothesized that Coastal Plain accessions would have different seed weights than the Ohio River Valley accessions.
Materials and Methods
This study consisted of a mapping exercise, seed weight data collection, and then analysis. The mapping exercise discerned the positions of Coastal Plain and Ohio River Valley accessions using the geographic Fall Line as the demarcation between Coastal Plain and Ohio River Valley accessions. Accessions found on the corresponding side of the Fall Line were assigned to a Coastal Plain or Ohio River Valley. This mapping was conducted in ARCGIS software.
Seed weight data for a handful of accessions was accessible from the USDA GRIN system. Notably, these seed weight data overrepresented Coastal accessions rather than Ohio River Valley accessions. Additional data came from Quesenberry et al (2003) as well as manual measurements. Quesenberry et al (2003) selected accessions from TX, GA, MS, and FL. One accession in this study came from the Ohio River Valley. Seed from single plant selections of Cincinnati and Clarks River accessions were measured on a lab balance (Bonvoisin scale).
Data Analysis
Seed weight was treated as a completely randomized design. Initial analysis at the state-level was considered to address potential accession grouping at local scales. Differences in seed weight between accessions from the Coastal Plain and Ohio River Valley were also considered to test other known differences in accessions (e.g. annual or biennial life history). All analyses were conducted in SAS v9.4 (SAS Institute, Cary, NC). PROC GLIMMIX coded for a simple one-way ANOVA.
Results and Discussion
The present study sought to present more trait data on Ohio River Valley accessions. Statistical analysis failed to detect significant differences in seed weights at the state level or between Coastal Plain and Ohio River Valley accessions. This result is surprising given other differences in life history, flower color, and growth habits among buffalo clover populations (Kubesch et al., 2022). Notably, Kentucky accessions are underrepresented in USDA GRIN databases. Only one Kentucky accession is presented in the current work. Even populations in Texas have similar seed weights. A limitation of the present study is that the Cincinnati and Clarks River accessions only represent single plants rather than the potential diversity present within each of those populations.
Table 1. Seed weights (g/100 seed) of buffalo clover (Trifolium reflexum) accessions across the eastern United States.
In comparison to Quesenberry et al., 2003 as well as the results of this work, the global average buffalo clover seed weight is ~ 0.085 g/100 seed. This rate corresponds to ~ 534000 seeds/lb. In comparison to white and red clover frost seeding rates, a buffalo clover would be ~2-3 lb/A (Kubesch et al., 2020). This work could support future work regarding seeding.
This work provides a baseline for developing seeding rates for restoration contexts. Seeding rates for introduced forage crops, such as tall fescue (Schedonorus arundinaceus) or red clover seek to produce approximately 1000 plants per square meter in a pure stand and adjust this rate to correspond to proportion in a mixture (Ohio Agronomy Guide, 2017) (David Barker, personal communication). Using this seeding target, a table of rates based on the average seed weight for buffalo clover can be calculated assuming an 85% germination rate (Table 2). Clover proportions in a pasture stand ideally range 20-40%, and so a 2-4 lb/A rate appears appropriate to correspond to seeding a mixture. Similar rates might work in more complex native seed mixtures.
Table 2. Seeding rate (lb/A) for buffalo clover (Trifolium reflexum) in pure stands and in seeding mixtures. Seeding calculations derived from and presented similarly to Ohio Agronomy Guide (2017).
| Seeding Proportion | Pure | 3/4 | 1/2 | 1/3 | 1/4 | 1/8 |
| lb/A | 9 | 7 | 5 | 3 | 2 | 1 |
Seed dormancy mechanisms are not directly addressed in this work. Notably, some accessions germinate without scarification while others require some removal of the seed coat in order to germinate (Kubesch et al., 2022). This issue can be overcome by adjusting the seeding rate to account for germination after scarification.
Mathematically:
Seeding rate desired * (100 seeds planted / germinated seeds after scarification) = seeding rate needed
Conclusions
This study is one of the first records of seed weights for buffalo clover (Trifolium reflexum) in the academic and technical literature. The current survey suggests that there are no differences in seed weights among accessions from the Ohio River Valley and the Coastal Plain, though differences may exist when additional traits are included.
For accessions with limited seed dormancy, buffalo clover might be seeded into agricultural grasslands and existing natural areas at a 2-4 lb/A rate. Optimizing seeding rate is worthy of further study. Future research should assess how buffalo clover establishes relative to red and white clovers. Additional pot and field studies need to be conducted to determine whether seeding rates should be further altered to account for establishment challenges.
Acknowledgements
This manuscript improved with suggestions from two reviewers. Kubesch wishes to thank his wife, Sarah Grace, who encourages him to conserve native clovers. This research was supported by Drs. John Fike and Ben Tracy, John James, and Velva Groover at Virginia Tech.
References
References for this article are hyperlinked throughout for ease of access. If any links fail to open, please contact the corresponding author: Jonathan Kubesch, M.S. (jakubesch@gmail.com).