PDF version
for printing

© 2005 Plant Management Network.
Accepted for publication 3 August 2005. Published 6 September 2005.

Twin-Row Corn Production: An Evaluation in the Mid-Atlantic Delmarva Region

Robert J. Kratochvil, Extension Specialist, Grain and Oil Crops, Department of Natural Resource Sciences and Landscape Architecture, University of Maryland, College Park 20742; and Richard W. Taylor, Extension Agronomy Specialist, Department of Plant and Soil Sciences, University of Delaware, Newark 19716

Corresponding author: Robert J. Kratochvil. rkratoch@umd.edu

Kratochvil, R. J., and Taylor, R. W. 2005. Twin-row corn production: An evaluation in the mid-Atlantic Delmarva region. Online. Crop Management doi:10.1094/CM-2005-0906-01-RS.

Abstract

Corn (Zea maize L.) produced in twin rows (two rows 7.5 inches apart on 30-inch centers) was compared to corn produced in rows spaced 30 inches apart over a range of plant populations at three Maryland locations and one Delaware location during 2003 and 2004. Grain yield averaged over the hybrids and populations differed between the two row spacing treatments at only one of the four locations and in that instance yield was greater in 30-inch rows. Corn produced in twin rows did not produce significantly better yields as plant population increased from 24,000 to 32,000 plants per acre. Only one of twenty-eight hybrid by row spacing combinations evaluated across the four locations had a significant linear response to population for yield. Yield response to population for both row spacing treatments varied across locations and hybrids. Greater levels of lodging occurred as population increased for twin-row corn. A greater incidence of stalk rot was also observed in twin-row corn. These results did nothing to indicate that planting corn in twin rows instead of the conventional 30-inch rows in the Delmarva region would be a beneficial management practice change.

Introduction

A variety of narrow row orientations for corn have been evaluated with the yield benefit varying and frequently dependent upon where the testing was conducted (3,4,5,6,7,8,11). Row-spacing studies in Maryland first compared 15- to 30-inch row spacing (7) and later, 20- to 30-inch rows (unpublished data). Those studies determined that producing corn in 30-inch rows remained the best option. The latest narrow-row corn promotional effort has been for twin rows. The patenting of a Precision Seed System by Great Plains Manufacturing (Salina, KS) allows their grain drill to plant numerous crops including corn. Because the drill has 7.5-inch spacing between planter units, different row spacing configurations are possible including twin rows that can be harvested with a standard, 30-inch corn head as well as accommodate post-emergent herbicides and side-dress nitrogen applications. The manufacturer has promoted the Precision Seed System for twin-row corn planting. Farmer testimonials indicate twin rows improved yields from 6 to 25 bu/acre over 30-inch rows. Contrary to these unverified reports, Becks Hybrids evaluated six hybrids in twin and 30-inch rows during 2003 (1). Five of the hybrids produced greatest yield in 30-inch rows. One that had erect leaf morphology produced more corn (2.5 bu/acre) in twin rows (no indication of significance). Universities have reported varying results for twin-row production. Over three years in Ohio a 4 to 15% advantage for twin rows compared to single rows was observed (10). Reports from Pennsylvania (9) and Mississippi (2) indicated no significant differences in yield between twin and single row production. Minimal use of twin-row corn production has occurred in Maryland and Delaware. This study was designed to increase knowledge about twin-row corn production for the Maryland and Delaware region. It had two primary objectives:

• Evaluate the performance of corn hybrids (grain system) produced in twin rows compared to 30-inch rows; and

• Compare both row spacing arrangements over a range of plant populations.

Corn in this study was planted at three locations during 2003, at the University of Delaware Research and Education Center at Georgetown on 25 April, at the University of Maryland’s Lower Eastern Shore Research and Education Center at Poplar Hill on 25 April, and at the University of Maryland’s Wye Research and Education Center on 30 April. During 2004, corn was planted only at the Wye Research and Education Center on 6 May. The experimental design was a split-split plot treatment arrangement using a randomized complete block with four replications per location. Whole plots were four hybrids in 2003 and two in 2004, split plots were row spacing treatments, and the split-split plots were three populations during 2003 and six during 2004.

Each twin-row plot had six sets of twin rows (two rows 7.5 inches apart on 30-inch centers) that were planted with a Great Plains 1510 drill equipped with precision seed meters. In order to plant the plots as quickly as possible, the 30-inch row plots were planted using a John Deere planter located at each site. Four non-GMO hybrids representing two seed companies were used during 2003. NK brand N70-D5 (111 day relative maturity (RM)) with a semi-flex ear and erect leaf morphology, NK brand N65-M7 (109 day RM) with a flex ear and lax leaf morphology, Augusta Seeds 4567 (113 day RM), and Augusta Seeds 2067 (90-day RM) were used. Only the two NK brand hybrids were planted during 2004. Plots were planted at a population rate that exceeded the highest treatment population by at least 10% and then thinned to the treatment populations (24,000, 28,000, and 32,000 plants per acre) approximately five weeks post-planting. Three additional populations (30,000, 35,000, and 40,000 plants per acre) were included during 2004. No seedling emergence measurements were made, however, when the plots were thinned none had less plants than the treatment population. Additionally, plants in both row spacing treatments were at the same vegetative growth stage. These two outcomes for seedling establishment indicated that the use of the two different planters had no effect on either seed germination or rate of seedling establishment in the plots. Standard fertility and weed management practices were used. A nitrogen rate of 160 lb/acre was applied at planting.

Stalk size was measured during 2004 at growth stage R5 using an electronic caliper. Stalk diameter for 10 consecutive plants was measured at the widest point on the elliptical stalk at the first internode above the brace roots. Just prior to harvest, lodging ratings (percent lodged plants) were assessed by counting the number of plants in the two center rows that were broken below the ear and/or leaning greater than 45°. Lodging causes were not identified. Stalk rot assessments were made by exerting pressure between thumb and finger to the first internode above the brace roots for all plants in one of the center rows. Stalk rot was determined present if the pressure easily indented the stalk. Plots were harvested 17 September at Georgetown, 24 September at Poplar Hill, and 6 October both years at Wye using a Massey Ferguson 8-XP plot combine (Kincaid Equipment Manufacturing, Haven, KS) equipped with Harvestmaster grain gauge system (Juniper Systems Inc., Logan, UT) that measured grain weight and moisture content. Grain data was converted to 15.5% moisture content for statistical analyses.

Data for the two years was analyzed separately because of the changes in treatments for 2004. Analyses of variance were conducted using PROC Mixed (SAS Institute Inc., Cary, NC). Since a number of significant interactions between locations and the other variables occurred in 2003, the analyses of variance were conducted by location. For these analyses, replication was assigned as a random variable, hybrid and row spacing were fixed variables, and population was handled as a quantitative variable. Mean separation analyses for row spacing effects was done using LSD when significance was indicated using Fisher’s protected F test (P ≤ 0.05). Regression analysis (PROC Reg) was used to evaluate the population effects. This analysis was conducted by location and hybrid to determine if a linear or non-linear response to population occurred for the two row spacing treatments and to assess if the response to population was different between the two row spacing treatments for each hybrid.

Row Spacing Effect on Yield

At three of the four locations during the two years, row spacing had no effect on yield. However, at Poplar Hill during 2003 there was a significant yield difference. The 30-inch rows produced more corn than was produced in twin rows (Table 1). There were no hybrid × row spacing interactions at any of the locations either year indicating that hybrids responded similarly to row spacing. Since there was no interaction between hybrid and row spacing observed either year, the different leaf morphologies (erect leaf versus semi-erect or lax leaf) that the two NK brand hybrids represented apparently had no influence upon twin-row yield. This lack of effect for leaf morphology on yield was contrary to the observation by Becks Hybrids (1).

Table 1. Grain yield by row spacing treatment and location averaged over the four hybrids and three populations during 2003 and two hybrids and five populations during 2004.

Population Effect on Yield

When corn seeds are planted to be spatially more equidistant in the field, the plants should respond better to available light, moisture, and nutrients and produce greater yield than when the plants are spaced closer together. Since planting corn in twin rows provides more equidistant plant arrangements compared with 30-inch rows, a significant yield increase was expected as populations increased. One indication that this occurred would have been a positive linear response for yield with increasing population. Of the 28-row spacing × hybrid combinations evaluated for linear response across plant populations during the two years, only one had a significant linear trend (P ≤ 0.05): NK N70D5 in 30-inch rows at Georgetown (Ŷ = 120.5 + 0.0022x; R² = 0.45*) (Fig. 1). The other 27 combinations had neither significant linear nor non-linear trends. Corn produced in twin rows did not produce better yield as population increased. In general, yield response to population for the two row spacing treatments varied by location and hybrid (Fig. 1).

Fig. 1. Yield response by location and row spacing to plant population for two hybrids, NK brand N65M7 (lax leaf morphology) and N70D5 (erect leaf morphology), evaluated in the twin-row corn production study conducted in Maryland and Delaware during 2003 and 2004. The only significant linear response to population observed in the study was for N70D5 in 30-inch rows at Georgetown. Locations: GT = Georgetown, PH = Poplar Hill, Wye = Wye; Row Width: TR = twin-rows, 30 = 30-inch rows; Year (03) = 2003; (04) = 2004. Note: Wye was the only location for the study in 2004.

Lodging

Lodging differences were observed across the locations for the two row spacing treatments during 2003. The least amount of lodging for any location was at Georgetown (4% lodged plants) primarily because the corn was harvested prior to Hurricane Isabella hitting the region on 18 September. There was no lodging difference between the two row spacing treatments at Georgetown, however there was significantly more lodging in twin rows than 30-inch rows (41 versus 32%) at Poplar Hill and (50 versus 40%) at Wye. Row spacing effects on lodging were consistent across hybrids and populations.

During 2003, lodging increased as plant population increased (Table 2) and more lodging occurred in twin rows than in 30-inch rows. While the magnitude of lodging was less in 2004 (3%) than 2003 across both row spacings, (Table 3) more lodging occurred for twin rows than 30-inch rows at 35,000 plants per acre. Increased lodging at higher plant populations was expected, but increased lodging for twin rows was not expected. The plants in twin rows had better spatial arrangement within and among the rows that should have allowed development of larger, stronger stalks and root systems and should have reduced lodging. The greater incidence of lodging in twin rows may possibly have been the reason there was no yield benefit observed for corn produced in twin rows.

Table 2. Lodging by location in 2003 for the three plant populations
averaged over the four hybrids and two row spacing treatments that
were used in the twin-row study.

 ^a Lodging means for a location followed by the same letter are not
significantly different at P ≤ 0.05.

Table 3. Lodging response by row spacing and population treatments averaged across the two hybrids used in the twin-row corn study conducted at Wye during 2004.

 ^a Lodging within a population and followed by the same letter are not significantly different at P ≤ 0.05.

Stalk Diameter

Typically, stalk sizes increase as plant-to-plant spacing increases within a row. In general, stalk diameter for corn in twin rows was 1 to 1.5 mm greater than for plants in 30-inch rows, as was expected. As plant population increased, stalk diameter decreased regardless of row spacing. The range in stalk size was 21 to 22 mm at 24,000 plants per acre to approximately 18 mm at 40,000 plants per acre. The change in stalk size as population increased explained the increased susceptibility to lodging that occurred both years as population increased. However, the increased stalk size of corn in twin rows compared with 30-inch rows does not explain why lodging was greater in twin rows. An additional factor must have been involved.

Stalk Rot

Stalk rot is caused by a number of fungal species including Diplodia, Gibberella, Fusarium, and Pythium spp. Stalk rot can be a major contributing factor to corn lodging. Stalk rot differences were observed among locations, hybrids, populations, and between row spacing treatments during 2003 but there were no interactions that confounded the results. During 2003, significantly more stalk rot was measured in twin rows (20%) than 30-inch rows (16%) but no significant differences were observed in 2004. The 4% additional stalk rot incidence observed for twin rows in 2003 was likely a contributing factor to the greater amount of lodging observed in twin rows. Additionally, stalk rot incidence increased with increasing population both years (Table 4), similar to the lodging response during the two years. Stalk rots frequently occur when the crop experiences stress. High plant populations are considered a predisposing factor for stalk rot so the response observed during 2004 for the three highest populations was not unexpected.

Table 4. Stalk rot incidence by population averaged over locations, hybrids and row spacings for the two years of the twin-row study.

 ^a Stalk rot means within a year followed by the same letter are not significantly different at P ≤ 0.05.

As with other narrow-row corn investigations conducted in this region, no significant grain yield advantage was observed for corn produced in twin rows. And there was no evidence to support the premise that if corn is grown in twin rows, the plant population can be increased in order to achieve a better yield than can be realized with 30-inch rows. In this study, response to population varied by location, hybrid, and row spacing treatment. The results observed in this study indicated that a seeding rate goal to achieve approximately 28,000 plants per acre should be suitable for attaining maximum yield for either twin or 30-inch row corn. There was an inexplicable increase in stalk rot incidence within twin rows accompanied by more lodging that increased as population increased possibly inhibiting the yield improvement potential that hypothetically should be attained with better plant spatial arrangement as would occur with twin rows. Since stalk rot causal agents are opportunistic organisms that attack when plants are under stress, such as infestation by European corn borer, the use of Bt hybrids may be better suited for twin-row production. Additional research is necessary to test this hypothesis. These findings for twin-row production of corn should not discredit the potential benefits of the precision seed system that allows a grain drill to plant numerous crops including corn in 30-inch rows. The opportunity for a farmer to have one planter for all planting needs is now a reality.

Acknowledgments

The authors thank the Agricultural Experiment Station staffs at both Universities for their assistance. The Maryland Grain Producers’ Utilization Board is recognized for its grant support for this project.

Literature Cited

1. Becks Hybrids. 2003. Becks twin-row versus 30-inch row corn population study – 2003. Online. 2003 research report for Becks Hybrids.

2. Buehring, N. W., Harrison, M. P., and Dobbs, R. R. 2003. Corn response to twin and narrow rows with selected seeding rates. Annual Report 2002 of the North Mississippi Research and Education Center. Mississippi Agricultural & Forestry Experiment Station Information Bulletin 398:43-49.

3. Cox, W. J., and Cherney, D. R. 2002. Evaluation of narrow-row corn forage in field-scale studies. Agron. J. 94:321-325.

4. Cox, W. J., Cherney, D. R., and Hanchar, J. J. 1998. Row spacing, hybrid, and plant density effects on corn silage yield and quality. J. Prod. Agric. 11:128-134.

5. Farnham, D. E. 2001. Row spacing, plant density, and hybrid effects on corn grain yield and moisture. Agron. J. 93:1049-1053.

6. Karlen, D. L., and Camp, C. R. 1985. Row spacing, plant population, and water management effects on corn in the atlantic coastal plain. Agron.J. 77:393-398.

7. Kratochvil, R. J., and Miller, T. 2002. Effect of row width and plant population on the performance of corn grown for grain in Maryland. Mid-Atlantic Grain and Forage Journal. Online. Rutgers University, New Brunswick, NJ. pp. 1-7.

8. Porter, P. M., Hicks, D. R., Lueschen, W. E., Ford, J. H., Warnes, D. D., and Hoverstad, T. R. 1997. Corn response to row width and plant population in the northern Corn Belt. J. Prod. Agric. 10:293-300.

9. Roth, G. W., Harkcom, S., Heinbaugh, S., and Antle, M. 2003. Comparison of twin-row and single row no-till corn planted for grain. Crop Management Research Report No. 02-01. Penn State University. Department of Crop and Soils Sciences. University Park, PA.

10. Thomison, P., and Geyer, A. 2003. Twin-row corn: An alternative to narrow row corn. The Ohio State University Extension Newsletter 03-05. March 4-17, 2003. Online.

11. Widdicombe, W. D., and Thelen, K. D. 2002. Row width and plant density effect on corn forage hybrids. Agron. J. 94:326-33