Optimum Row Width for Corn -- Revisited

Crop and Soil Environmental News, March 1998

Daniel E. Brann
Extension Specialist, Grains
Virginia Tech

A Virginia Tech study at three locations from 1966 through 1968 showed corn yields of 130, 132, and 133 Bu/A for 32", 24", and 16" rows, respectively at 22,000 plants/A. When plant population was increased to 28,000 plants/A, yields were 123, 130, and 137 Bu/A for 32", 24", and 16" rows, respectively. Thus, in the 1960s study, narrow row width increased grain yields ONLY when the population was 28,000 plants/A. Also, yields were not increased when plant population was increased from 22,000 to 28,000 except at 16" rows and then the increase was relatively small. A companion study in the 1960s evaluating corn for silage showed silage yields of 17.6, 17.7, and 18.4 tons/A for 32", 24", and 16" rows, respectively at 22,000 plants/A. At 28,000 plants/A, yields were 17.3, 17.9, and 19.1 tons/A at 32", 24", and 16" rows, respectively. As with grain, silage yields were not increased by increasing population from 22,000 to 28,000 until the row width was reduced to 16".

The above research is 30 years old! Does it still apply? That question is being asked by corn farmers and researchers across the country. Narrow row (15-20 inches) planters for soybeans are already on many farms, narrow row headers are available, new hybrids with very high yield potential are available, double-crop corn for grain is being evaluated, and Roundup ready corn will be available to a limited extent in 1998. My counterparts in other states have already conducted research to help answer the question, but more work would certainly be beneficial since it is rather expensive to implement narrow corn rows.

The theory behind narrow row corn is that yields of any crop will be maximized when plants are uniformly distributed across the field. For example, at 30,000 plants/A corn plants would be 14" apart in 15" rows compared to 7" apart in 30" rows. Theoretically, the more equidistant spacing should minimize the effects of competition among plants, especially early in the season before a plant canopy is formed.

Grain Yield Response to Narrow Rows

Dr. Greg Roth, Associate Professor of Agronomy and Extension Specialist at Penn State University summarized research across the country as shown in Table 1. Corn yields generally increased with narrow rows in the mid-western "Corn Belt" and northern states such as Michigan and Minnesota. The yield increase from 22" rows (or less) compared to 30" rows was 2.7 to 4.5% in Indiana and Iowa and 7.3 to 8.8% in Minnesota and Michigan. However, studies in Kentucky and Tennessee have shown either similar yields of corn in 20 and 30" rows or a 4.3% decrease with 20" rows. The yield response to narrow rows seems to be greatest in shorter season environments and less or even negative in southern locations.

The Pennsylvania data where yields increased more with narrow rows at 34,000 than at 27,000 plants/A shows that the question of optimum row width should not be separated from plant population. Research by University of Kentucky Extension Specialists, Morris Bitzer, and Jim Herbek over a three year period showed that Pioneer Brand 3163 corn yield was influenced by both row width and plant population. However, in contrast to the Pennsylvania data, there was no significant row width by plant population interactions within any year or across years. The Kentucky data averaged over eight locations for three years as shown in Table 2.

The average yields at each Kentucky location in 1995, 1996, and 1997, respectively were: Hardin County, 184, 121, and 99 Bu/A, McLean County, 185, 174, and 176 Bu/A, and Hickman County (1996 and 1997) 215, and 203 Bu/A. Corn planted in 30" rows produced the highest yield of 175 Bu/A while corn produced in 36" rows produced the lowest yield of 169 Bu/A. Twenty inch rows produced an intermediate yield of 170 Bu/A. Statistically, there is only a 10% probability that the yield difference of six Bu/A between 30 and 36" rows was due to chance and thus under the conditions of these experiments there is a high probability that 30 inch rows increased yield over 36 inch rows. The difference of 5 Bu/A in favor of 30" over 20" rows is not significant at the 10% level of probability but would likely be significant at the 15% level. Thus, there is also a high probability that under the conditions of these experiments 20" rows produced less yield than 30" rows. Further work is needed to fully explain why the yield was lower at 20" rows. The most dramatic and consistent response in this experiment was a yield increase all three years at all locations when plant populations were increased from 20,000 to 26,000 plants/A. Yield increases with plant populations up to 30,000 plants/A occurred the first two years, but there was no yield increase in 1997 when plant population was increased from 26,000 to 30,000. Yields were generally above 175 Bu/A when the higher plant populations were favorable.

This article is primarily about row width, but I would like to interject a comment about corn responses to plant population. Several years of corn plant population research was conducted in the late 1980's and 1990's at Virginia Tech and Maryland. These studies have shown that final stands of hybrids with yield stretch potential (ability to produce one large or two normal ears) should be in the 18,000-20,000 plants/A on 80-120 Bu/A long term average soils. This will assure a yield potential of at least 150 Bu/A in excellent seasons. A 1993 through 1995 Kentucky study by Morris Bitzer and Richard Barnhisel evaluated Pioneer Brand 3140 and Dekalb 646 on soils with varying topsoil depth. The data showed that on a thin droughty soil corn yield decreases from 86 Bu/A at 18,000 to 85 Bu/A at 20,000 and only 77 Bu/A at 26,000 seeds dropped/A. On the medium depth soil corn yields were: Seed drop/A of 20,000 yielded 104 Bu/A, and a seed drop/A of 26,000 yielded 101 Bu/A. On the deep soil increasing seed drop/A from 20,000 to 26,000 increased yields from 123 to 129 Bu/A. Probable yield level must be considered when selecting optimum plant population! Planting at 25,000 seeds/A or higher should generally be considered only on soils with about four inches of available water holding capacity and long term corn yields above 125 Bu/A. As we implement precision farming principles and begin varying plant populations with soil productivity we will learn much more through on-farm research. For now, under Virginia conditions, I feel that plant populations above 22,000 plants/A at any row width are likely to be significantly beneficial only when probable yields exceed 125 Bu/A.

We plan to initiate row width, plant population studies both full season and double-cropped with corn in 1998. This, coupled with additional work from surrounding states will give producers information from which to make the best decisions about how corn should be planted on their farm.

Silage Yield Responses to Row Width

Research by Dr. Greg Roth, Associate Professor of Agronomy and Extension Specialist at Penn State University, evaluated 30" and 15" rows at a population of 32,000 plants/A in 10 experiments across Pennsylvania. The narrow rows increased yields by 8.8%. Similar studies in New York have shown about a 5% benefit to decreasing row spacing from 30" to 15" rows. Forage quality of corn silage produced in narrow rows has been similar to that produced in wide rows at the same plant population.

Table 1. Grain Yield Responses to Narrow Row Spacing at Multiple Locations in the United States
Location	Experiments	Narrow Row Spacing	Yield Response
		(in)	% Change
Michigan State	10	22	8.8
Minnesota	6	20	7.3
Purdue	9	15	2.7
Iowa State	5	20	4.5
Pennsylvania    27,000 PPA    34.000 PPA	2 2	15 15	3.2 11.0
Pioneer Hibred	14	224.0
Univ. of Kentucky	5	20	0.0
Univ. of Tennessee	3	20	-4.3
*Yield response compared to 30 inch rows.

Table 2. Effect of Row Width and Plant Population on Corn Yields in Eight Kentucky Tests From 1995-1997.
Row Width/Population	Average Yield
	Bu/Acre
20 inch	170ab*
30 inch	175a
36 inch	169b
20,000	164a
26,000	171b
30,000	178c
* 0.10 level of significance for row width; 0.05 level of significance for plant population.

Visit Virginia Cooperative Extension