is part of the Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

# How Much Would You Spend to Learn?

Trials set up like scientists want, mean losing on some land.

Suppose you're planting 176,000 soybean seeds per acre because that's what you've done for years, and that's where your drill is set. The local county Extension ag educator says you could drop back 20,000 seed per acre and save \$7 to \$8 per acre in seed costs. Your answer is simple - prove it!

He answers back that he can prove it if you will put out a test plot. He says it can even be a replicated strip trial instead of a small plot, which tends to take more time. You're about ready to say yes, thinking you're only doing two rates, when he says, no, we need at least five rates, and one of them needs to be low, like 55,000 to- 60,000 seeds per acre.

Wait a minute, you say. You know the yield won't be as high on those strips, and you know you'll give up more yield than you save on seed. Why is it necessary to go so far to prove something when you already know it won't work? You're not too keen on the next suggestion either - 90,000 seeds per acre. And the top rate - 275,000 - he wants you to waste seed on purpose because you know you don't need that many beans out there.

Dave Smith, Extension ag educator in Johnson County, helped get Purdue's new emphasis on on-farm trials rolling last season. He worked with one set of farmers on a twin-row study and with two other farmers on a nitrogen rate study. He didn't ask them to go to zero on N application rate as one treatment. He realizes there's a cost with losing yield.

The problem, Smith says, is that you need enough points to prove that your hypothesis, that you can cut back on seeding rate but only so far, is right. If you don't go low enough, you won't find the tip in the line where yields finally start to fall off. If you don't go high enough, you won't know for sure that higher populations don't help bump up yields.

It's an argument you won't win with every farmer, Smith realizes. In a perfect world, you would do five or six rates and drop very low and very high to set the extremes. That's what most researchers intend to do in their trials. However, if the plot covers 40 acres to get the reps in, and you know you're taking at least a three-bushel hit on 20%of those acres, or 8 acres, for example. That's a \$40 per acre hit this year, or \$320 for the field.

Will what you learn be worth it? It might if the conclusion is you could drop 20,000 per acre fewer seeds than your current rate next year and do just as well. At \$7 per acre seed savings, that's \$5.600 in savings for a farmer with 800 acres of soybeans.