Crops: The Lynch File - How many soy seeds per acre do you need to plant?

Thursday, April 2, 2009

by PAT LYNCH

For years, there has been open disagreement on what seeding rate you should use for soys. There has been some consensus that seven-inch rows require more seeds per acre than 14- or 15-inch rows. But after that there is little general agreement. Some of us have been advocating lower populations than others. All of us are using our own experiences to come to the best seeding rate. The seeding rate being put forward by the Ontario Ministry of Agriculture, Food and Rural Affairs this past couple of years is lower than what was once recommended. This is due to the hard work of Horst Bohner who has spent a lot of time doing on-farm trials on soybean seeding rates.

The real issue comes down to the belief that "we do not have a good piece of equipment to plant soys." You get to choose between a drill which has seen little change in the last 20 years and which is basically designed to plant cereals at 1.4 million seeds per acre, or a planter which is designed to plant about 30,000 seeds per acre.

So, in one case, the equipment is designed to drop seed at about eight times the rate we want to drop for soys. The other equipment planter is designed to drop seed at about 15 per cent of the rate we want to drop soys seed. So it is no wonder we end up with terrible seed drop and placement.

There is a lot of research to suggest that all you need is 100,000 plants to get top yields. To get this type of population, all we really need to do is drop maybe 130,000 seeds per acre. But, because of poor seed placement and poor seed metering devices, you often have to drop almost twice that rate to get maximum yields.

This point was really driven home last year when a grower commented that, when he watched the seed drop on his drill, he was getting a 25 per cent difference in seed drop between rows. We would never accept that with corn. Imagine dropping corn seed when the best you could do was a 25 per cent variation between rows.

So if you were averaging a 30,000-seed drop, you would have one row dropping 26,250 and another row dropping almost 34,000 seeds per acre. Then, imagine seeding depth being so inconsistent that you ended up with an average population of 25,000 plants per acre and a range of 34,000 to 14,000.

It's ridiculous, but that is what the drill does.

Figure 1 has actual populations from what I believe is a very good grower using a well-maintained and fairly new drill. I have estimated the seeding rate that was used to get the various populations. Each estimated seeding rate is represents the average of quarter of the populations. In other words, when we did populations, they were divided into the top 25 per cent, next 25 per cent, and so on. The grower dropped an average of 185,000 seeds per acre. We used standard population techniques to measure plant populations.

So what can you do about this? You can do your own population study this year and check for variation in plants per acre between rows. Then you can go to the equipment supplier and ask if they can do better. If you had the variation from row to row with corn plants that you have with soys, someone would do something. Put pressure onto the equipment folks to give us a better system to plant soys. BF

Pat Lynch CCA (ON) is head agronomist for Cargill in Ontario.