Frustrated by a lack of information about soybean varieties for his high-pH soils, Ted Klug decided to generate his own data.

"We have very limited information on soybeans in this area," says Klug, who grows 500 irrigated acres of corn, soybeans and alfalfa with his son, Ted Jr., near Maxwell, NE.

Before 1998, the Klugs had grown soybeans for five years in only neutral-pH soils. In '97, their yields peaked at 74 bu/acre. Last spring, they decided to up their soybean acreage.

"We plan to start growing popcorn or food-grade white corn and those hybrids must be grown on ground that's clean - and there's nothing cleaner than soybean ground," explains Klug, a devout ridge-tiller.

To expand their soybean acreage, the Klugs will have to utilize high-pH soils; however, they were unsure which varieties would be best-suited for those soils.

In a search for some answers, they decided to conduct their own on-farm research. Working with agronomists from the Farmland Service Cooperative in Gothenburg, and their Pioneer sales rep, the Klugs planted three test plots. The company reps selected and provided the varieties to be grown and assisted with planting. Klug kept meticulous records, noting variety number, seed size, seeds per acre planted, etc.

At each of the three sites, six rows of each variety were planted in 36" rows. All planting was done in early to mid-May.

At the first site, with a soil pH of 8.2, they planted 11 varieties from Novartis and Asgrow. Yields on the five acres ranged from 47 to almost 61 bu/acre. The top yielder was S30-06.

At the second location, five varieties were planted on one-fourth acre in soil with a pH of 8.1. Yields ranged from 42 to 58 bu/acre. Asgrow A2833 was the top yielder.

At the third site, in neutral-pH soil, the Klugs planted 14 varieties from Asgrow, Pioneer and Golden Harvest. The winner was Asgrow 2553 with 66 bu/acre. The average of the 16-acre site was 52 bu/acre.

"Two heavy doses of hail took its toll on that field," says Klug.

The Klugs will use the information gathered in last year's tests to make this year's variety selections.

They'll plant the top yielders from the first two sites on several acres of high-pH soils. And they'll plant more test plots.

"I'll keep trying," he says. "Data from several locations over multiple years is far more useful than data from a single location."

However, the father and son will change the way they run the tests.

"I'll talk to several seed company reps and ask only for soybeans that are suited to high-pH soils. Then I'll only buy the top one or two varieties from each company."

They'll also plant up to 250,000 seeds/acre in each test plot - a strategy they've used with success on their other soybean acres.

"Our yields jumped from 55 to 74 bu/acre when I increased the seeds we planted from 180,000 to 250,000 per acre," Klug reports. "And that's the only thing I did differently from one year to the next."

Based on a $15 per bag seed cost, Klug figures it only costs about $7.63 more per acre to plant 250,000 seeds vs. 180,000.

He's confident that his own test results will help him make better decisions and bigger profits.

The Klugs are not alone in their quest for reliable results from on-farm variety tests.

In a University of Wisconsin survey, corn growers were asked what sources of information were most useful to them in choosing corn hybrids. Their answers: results of their own yield tests, corn company tests on their farm, test results close to their farm, university tests and information from corn company agronomists.

Soybean growers would probably give similar answers if asked the same question, believes Dale Hicks, University of Minnesota extension agronomist.

Other surveys also show that growers believe test results from their farms or locations close by are important for choosing top varieties to grow, Hicks says.

He says growers should also check out multiple-location test results from their state universities, if available. An extensive research project at the University of Minnesota, he explains, showed that multiple-location test results are the very best predictor of soybean yield performance.

That confirms Klug's earlier statement that considering data from several locations over multiple years is superior to results from a single location.