The author uses wooden markers to record the sowing dates and varieties of Swiss chard. At the end of the season, these markers serve as a reminder of what she did. (Lynn Karlin)

When you are a beginning gardener, a lot of what you learn is by trial and error. And when you have been gardening for many years, the exact same thing is true.

First, let’s consider trials (as in evaluations, not ordeals). At an agricultural research center or experiment station, trials are what they do. A number of varieties are planted and their performance is observed and recorded. The same plant might be grown under different conditions and the results compared, using criteria such as yield, heat tolerance and taste. With taste, results are subjective, so multiple testers are employed. There’s also an element of chance. Hard science softens a bit when you’re dealing with factors such as weather, and even in rigorous academic trials, you might find an inconvenient hailstorm or long rainy spell noted as skewing certain results.

I like to think of our farm and home garden as our own experiment station. We’re always trying new varieties to see whether a certain spinach is slower to bolt, or whether one paste tomato makes a richer sauce than another we’ve been growing. Sometimes we’ll do a trial by sowing a number of varieties of one crop. It was by growing six kinds of Swiss chard that we found Argentata to be the hardiest for cold weather.

The same goes for our practices. If you plant cabbages closer together, will the heads be of a more convenient size? Yes. Does tilling bark mulch into the soil improve plant growth? Definitely not; nothing grew.

Try designing a controlled experiment in which you do something two ways. Want to know whether aerating the soil between crops with a broadfork (or digging fork) is worth doing? Loosen the soil in half of the bed, then sow a crop in the whole bed and compare the results.

With any trial, you’ll learn more if you keep records. We’ve found that 12-inch-long wooden plant markers, which we buy in a box of 100, are useful. (They’re available at Jung Seed, Peaceful Valley, Johnny’s Selected Seeds and more.) Put the sowing date of a crop on the label with a permanent black marker, along with the name of the plant. Save all of them at season’s end, bundled with a rubber band, as a record of what you did.

A notebook is a big help, too, even if it’s smeared with mud. Just remember to use a waterproof pen, and don’t use a red one. I found it fades quickly in the sun.

Often the most instructive part of trial-and-error gardening is the error part. For example, this year we planted our peas but didn’t get around to putting up a trellis to support them. Realizing that the right moment had passed, we consulted the catalogue from which they were chosen, and it turned out that all the varieties could be grown “with or without support.” Although even short-vined peas look tidier and are easier to pick when trellised, we found that letting them flop on the ground was not a disaster at all. I found that I could gently lift a handful of vines and probe them with my other hand. I could tell by feel which ones had filled their pods and which ones were rough-textured and therefore too old to bother with.

Another instructive mistake was the planting of an herb called papalo, popular in the Mexican province of Puebla, that a friend had suggested. “I taste cilantro, watermelon rind, mint and arugula all mixed together — maybe a little raw onion, too,” he raved. When mine had grown, I wrote him that to me it tasted bitter and unpleasant. He admitted that “it has a nickname that equates to ‘donkey breath’ because the flavor and scent are strong, and linger.” But I guess I had to learn the hard way.