No. 4 Article 9/April 29, 2010

The favorable April weather allowed corn planting at a pace that got growers off to an excellent jump-start for spring activities, allowing them to prepare for soybean planting ahead of "normal" schedule. One of the last decisions a grower will make before planting soybean seeds is how many to drop in the ground. This decision has become more economically critical as seed prices have increased with improved varieties, the addition of biotechnology traits, and the increasing popularity of seed treatments.

Soybean plants have the ability to compensate for a wide range of competitive pressures from surrounding plants. If plants are in a densely planted stand, they will only grow as a single primary stem and set pods on main nodes only. However, when the stand is less dense, the plants will "bush" by growing secondary branches, and pods set on these branches can account for significant seed yield. Soybean seeding rates in Illinois have typically ranged from 150% to 200% of the number of plants needed at harvest (75,000 to 100,000 plants/acre) to maximize yield. High seeding rates help insure against conditions that reduce soybean emergence. Historically, the cost of soybean seed was a relatively minor expense in the cropping operation, so the practice made a lot of sense, but today's higher seed costs have increased interest in reducing seeding rates to maximize economic return.

Data were compiled from three different experiments representing 50 site-years of data. These site-years cover many locations throughout Illinois, several growing seasons between 1998 and 2009, multiple row spacings (7.5, 15, and 30 inches), and several different fungicide seed treatment applications. In all cases, data were balanced and interactions were nonsignificant, so pooling the data to analyze for the response of yield to seeding rate was statistically valid. The results show the rather wide variation around an expected yield given any rate in the range of 50,000 to 200,000 seeds per acre over a wide range of conditions. Most importantly, the variation of the data is constant across the range of seeding rates, meaning that there is just as much yield variation at the high seeding rates as the low ones, which suggests that not that much greater "insurance" is provided by higher rates.

These data were also modeled to a quadratic equation to find the seeding rate that maximized yield, which occurred at approximately 200,000 seeds per acre (**Figure 2**). This makes sense regarding why growers have historically planted such high seeding rates. However, the yield response is slow to increase over a wide range of seeding rates, and that model can also be used to find economic optimum seeding rates by using the price of seed and the expected value of your crop.

*Figure 2. Soybean yield data for seeding rates ranging from 50,000 to 200,000 seeds per acre collected from 50 site-years from 1998 to 2009 throughout Illinois including drilled (7.5-in.), 15-in., and 30-in. row spacings and different fungicide seed treatments.*

Using that regression equation, I've produced three graphics that demonstrate how economic seeding rates have changed over the last 15 years based on this model. The seeding rate from 50,000 to 230,000 is in the left-hand column, and the other columns represent the partial expected return given the changing costs of seed in $/1,000 seeds at three soybean values: $8/bushel (**Figure 3**), $10/bushel (**Figure 4**), and $12/bushel (**Figure 5**). You can see that as seed costs have increased from below 15 cents per 1,000 seeds 15 years ago to nearly 45 cents per 1,000 seeds today, the economic seeding rates have decreased. Where a "recommendation" should remain under today's economic environment can be pretty wide given a number of conditions.

*Figure 3. Partial expected return based on quadratic yield response of 50 site-years of seeding rate data and various seed costs in $/1,000 seeds at $8/bushel soybean.*

*Figure 4. Partial expected return based on quadratic yield response of 50 site-years of seeding rate data and various seed costs in $/1,000 seeds at $10/bushel soybean.*

*Figure 5. Partial expected return based on quadratic yield response of 50 site-years of seeding rate data and various seed costs in $/1,000 seeds at $12/bushel soybean.*

The best way to determine a desired seeding rate is to understand your operation, including seedbed preparation, the accuracy of your planting equipment, and the quality of your seed. Perhaps the best rule of thumb is that a minimum of 100,000 uniformly spaced **plants** per acre can maximize yield for a very wide range of conditions. So you can determine how many seeds you should start with by dividing your desired plant stand establishment by the percentage of viable seeds planted, then dividing again by the percentage of your expected emergence rate. If you have a history of scouting your plant stand establishments following planting, that should give you knowledge of the expected percent stand establishment to use specific to your equipment, seedbed preparations, and soil/planting conditions. One factor that should be in our favor this year is the excellent start to the season, which will hopefully allow everyone to plant in conditions more conducive to maximizing plant stand establishment rates.*--Vince M. Davis*