As I prepare to embark on the largest variety trial I’ve ever attempted, I decided to seek some wisdom from Edmund Frost. Edmund’s farm is in Virginia and he’s a well known seed saver, experimental farmer and cucurbit breeder. Edmund runs a small seed company called Common Wealth Seed Growers.
We are seed growers, creating and growing a vibrant regional seed system in the mid-Atlantic and Southeast. We sell regionally-adapted, open-pollinated, organic seeds for a limited number of great varieties. We grow all the seeds we sell, and we only sell what grows well here. We specialize in Downy Mildew resistant cucurbit seed stocks.
Edmund has done large scale squash trials (among many other projects) and I asked him what his top 3 trialing tips would be. He’s an awesome seed breeder, but he can’t count:
- Make the planting as uniform as possible, in terms of soil type, drainage, rotation history and fertility. You may want to avoid field edges as they can have compaction issues. When I’m doing downy mildew trials it can be important to do an edge planting. The prevailing wind is from the west, so the plants on western edge are exposed to fewer DM spores blowing over them. So its important to at least do an edge planting on the west side. I don’t know if there is any phenomenon like this that could necessitate an edge planting for your trial – I can’t think of any but also don’t know okra well.
- Replication can be helpful. I have seen a variety do well in an observation trial that turned out to be a fluke. Usually you get the information you need without replication though. If you want to be able to do statistical analysis, you’ll need 3-4 replications, but I wouldn’t do that unless you have to for a grant or a publication. For my own informational purposes I would like to be able to do two replications of things but often can’t due to time and budget constraints. If you do replications, keep the planting uniform within each replication block. Its ok to vary from one block to the next.
- Different maturity dates can make it harder to compare varieties. For an example with okra: if one variety is coming in when its cool and wet, while another starts producing two weeks later when its hot and dry, it could be hard to compare tenderness. I don’t have an overall solution to this. For the example I gave it would help to keep it well watered. That is, unless you want to test drought tolerance. Another approach might be to vary planting dates based on expected days to maturity. But I wouldn’t necessarily recommend this for the first trial, or if you do don’t use more than two or three planting dates.
- Organic seed alliance has two publications on trialing, one from 2007 and one from 2018. I haven’t read the 2018 one yet, but the 2007 one was very helpful to me getting started with trials. Don’t worry if you aren’t able to meet their recommendations for plot sizes and number of replications. https://seedalliance.org/all-publications/
- One helpful piece of advice I got from OSA was – don’t try to assess more traits than you can handle. It might cause you to abandon the record keeping half way through. I guess this has been common with busy farmers who also take on variety trials and then get overwhelmed by the rest of the farm. I’ve generally evaluated more traits than they recommend, but their advice probably helped me do less than I would have.
- Make a plan to deliver a pretty short and digestible report on the trial, or at least one version that is short. A lot of people don’t want to read long and complex things.
My okra variety trial is going to focus on a limited set of observable traits over a wide number of varieties. My hope is to identify varieties that are worth exploring further and deeper with more extensive trials and replicated plots.
It’s going to be exciting and, thanks to Edmund, not quite so terrifying.