Ohio State University Extension Fact Sheet

Horticulture and Crop Science

2001 Fyffe Court, Columbus, OH 43210-1096

Performance of TopCross® High Oil Grain Production Systems for Corn, 1995-1997

AGF-133-98

Peter R. Thomison, Extension Agronomist, Hort. & Crop Sci.
Allen Geyer, Research Asst., Hort. & Crop Sci.
Tammy Dobbels, Ext. Agt., Agric. & Nat. Res. - Logan Co.
Tom Krill, Ext. Agt., Agric & Nat. Res. - Van Wert Co.
Larry Lotz, Ext. Agt., Agric. & Nat. Res. - Fayette Co.
Howard Siegrist, Ext. Agt., Agr. & Nat. Res. - Licking Co.

Corn hybrids commonly grown in the Corn Belt have a kernel oil concentration of about 4% on a dry weight basis. Specialty high oil corn hybrids contain 6% or more oil on a dry weight basis. High oil corn is attractive as a livestock feed because it has greater energy than normal corn. The caloric content of oil is 2.25 times greater than that of carbohydrates (which includes starch - the major form of carbohydrate found in corn kernels). Feeding trials with high oil corn generally indicate it has improved feed efficiency and leads to an increased rate of gain over conventional corn.

Commercial high oil corn hybrids have not been widely used by growers because their grain yield potential is lower than normal dent hybrids. Recently there has been interest in producing high oil corn using the TopCross® grain production system. The TopCross system may minimize the yield disadvantage associated with high oil single cross hybrids while enhancing grain nutrient composition. The TopCross high oil grain production system involves planting a blend of two types of corn. One type, representing 90 to 92% of the seed in the blend, is a hybrid that is designated as the "grain parent." The second type, representing 8 to 10% of the seed, is a special "pollinator." The grain parent is a male sterile version of an elite hybrid that may be in commercial production. The pollinator is a special line, available from Optimum Quality Grains, L.L.C. and licensed to seed companies, that sheds pollen within a TopCross grain production field. The pollen shed from these pollinator plants contain special genes that cause a kernel to produce a much larger than average germ or embryo (commonly called xenia effect). Since most of the oil and essential amino acids are in the germ, the oil, and thus the energy level, and protein quality of the grain produced by fertilization with these pollinators is enhanced. Pollinator plants contribute little to overall grain yield, but use resources such as soil nutrients, water, and sunlight. Their primary function is to provide pollen to the male sterile grain parent.

The blends used to produce TopCross high oil grain are designated TC Blend® seed corn products. Testing high oil TC Blends® is difficult due to isolation requirements. If pollen from normal, low oil corn hybrids pollinates male sterile hybrids in the blend, then the high oil trait would not be expressed. Although TopCross high oil production has increased from 50,000 acres to over 1 million acres during the past five years, limited university and Extension information is available on the agronomic performance of these new high oil TC Blends or the nutrient composition of TopCross grain.

In 1995-1997, field tests were conducted to compare the agronomic performance of a set of high oil TC Blends and their normal grain parent checks in different production environments across Ohio. In 1995-1996, Pfister Hybrid Corn Company provided five TC Blends (SuperKernoils[TM]) and their respective male fertile, low oil grain parent checks. These TC Blends contained Pfister pollinator "2." In 1997, Pfister Hybrid Corn Company provided 10 TC Blends and their respective grain parent checks. Five of the ten TC Blends contained Pfister pollinator "18" and the other five SuperKernoils contained Pfister pollinator "19." Pioneer brand hybrid 3394 (P3394) was also included as a check hybrid because of its high grain yield potential, adaptability, and popularity. Based on previous results from the Ohio Corn Performance Trials (which have compared commercial corn hybrids for oil and protein content), we knew that grain of P3394 was characterized by average oil content.

County Strip Plot Study

Strip plots were established at sites in Clark, Fayette, Franklin, Logan, Licking, Madison, Van Wert, Wayne, and Wood Counties in 1995-1997. In 1997, the five TC Blends containing Pfister pollinator 18 were evaluated at five sites; the five TC Blends containing Pfister pollinator 19 were evaluated at seven sites. A high oil single cross hybrid, Pioneer brand 34M55 (P34M55) was also included in the 1997 multi-county strip plot evaluations. Each TC Blend and single cross hybrid was planted in strip plots at least four rows wide (30-inch row spacings) and 200 feet in length.

Replicated Plot Study

The same set of TC Blends and grain parent check hybrids compared in county strip tests was also planted in a replicated trial at the OARDC Northwest Branch Research Farm at Hoytville in Wood County in 1995 and 1996. In 1997, the set of five TC Blends with Pfister pollinator 19 and their grain parent checks were planted in replicated trials at two locations - the OARDC Northwest Branch Research Farm and the OARDC Western Branch Research farm at South Charleston (S. Charleston) in Clark County. A high oil single cross hybrid was included in the study each year - Countrymark 652HO (CM652HO) in 1995-1996 and P34M55 in 1997. Each TC Blend and grain parent check plot consisted of four 30-inch rows, 200-feet in length. The high oil single cross hybrids were planted in twelve 30-inch row plots, 200 feet in length. The experimental design was a randomized block with five replicated blocks in 1995 and three replicated blocks in 1996-1997.

To minimize possible pollen contamination from neighboring male fertile check hybrids as well as any nearby normal corn, the TC Blend plots were planted in isolation approximately 200 feet from normal corn hybrids in 1995-1996. A 200-foot buffer separating the normal corn from the TC Blends was planted with excess TC Blend seed to minimize foreign pollen contamination. Plots were also planted with TC Blend seed as border (20-50 feet) on all sides of the isolation field. In 1997 the buffer separating the TC Blends and grain parent checks was reduced to 100 feet based on testing guidelines prepared by Optimum Quality Grains. The TC Blend test plots were planted adjacent to each other with a common pollinator. At all locations, the only nearby foreign pollen source was that of the male fertile grain parent plots.

Shortly before harvest, ten ears were randomly selected from plants in a 50-foot length of row in the center of each plot. These ears were shelled and a subsample of grain from each plot was submitted to the Ohio State University Grain Quality Laboratory (Wooster, OH) and Optimum® Quality Grains Laboratory (Urbandale, IA) for grain nutrient composition. Oil, protein, and starch content were determined by near infrared transmittance (NIRT) analysis. Non-ruminant metabolizable energy (M.E.) and lysine were estimated by calculation. Final plant stand, numbers of plants stalk lodged (stalk breakage below the ear), and barren or with nubbin ears were recorded at harvest. Plots were harvested by combine and grain yields were adjusted to 15.5% moisture.

Results

County Strip Plot Study

Grain yields of the TC Blends averaged across the five 1995 test sites (Fayette, Licking, Madison, VanWert, and Wood Counties) varied from 107 to 127 Bu/A, whereas yields of normal grain parents ranged from 116 to 136 Bu/A (Table 1). In 1996, TC Blend yields averaged across the five 1996 test sites (Fayette, Franklin, Licking, Madison, and Wood Counties) ranged from 115 to 134 Bu/A whereas normal grain parent yields ranged from 133 to 145 Bu/A (Table 2). In 1997, yields of TC Blends with pollinator 18 averaged across test sites (Franklin, Licking, Madison, Wayne, and Wooster) ranged from 115 to 146 Bu/A compared to normal parent check yields that ranged from 131 to 152 Bu/A (Table 3); yields of TC Blends with pollinator 19 averaged across seven test sites (Clark, Frankin, Logan, Licking, Madison, Wayne, and Wooster) ranged from 115 to 146 Bu/A compared to normal parent check yields that ranged from 131 to 152 Bu/A (Table 4).

To determine if variation in TC Blend and normal hybrid performance was significantly different, data from the five to seven county tests each year were combined and compared as a randomized complete block design with five to seven replicated blocks. A comparison of agronomic performance and grain composition from the combined data is presented in Tables 1 (1995), 2 (1996), 3 (1997 - pollinator 18 ), and 4 (1997 - pollinator 19).

In 1995, grain yields of the TC Blends were not significantly different from their respective grain parent checks (Table 1). Two of the five TC Blends produced yields similar to P3394. In 1996, four of the five TC Blends produced grain yields that were significantly lower than their respective grain parent checks but yields were not significantly different from P3394 (Table 2). In the 1997 multi-county evaluation of TC Blends with pollinator 18 (Table 3), one of the five TC Blends produced yields that were significantly lower than its respective grain parents; three TC Blends produced yields significantly less than 3394 but TC Blend yields were not significantly different from P34M55. In the 1997 multi-county evaluation of TC Blends with pollinator 19 (Table 4), two of the five TC Blends produced yields that were significantly lower than their respective grain parents; TC Blend yields were not significantly different from 3394; one TC Blend yielded significantly more than P34M55.

The average oil content was 2.9 percentage points greater in TopCross grain from TC Blends than in grain from normal parents and 3.3 percentage points greater than in grain from P3394 in 1995 and 1996 (Tables 1 and 2). In 1997 the average oil content was 3.3 percentage points and 3.7 percentage points higher in TopCross grain from SuperKernoils with pollinators 18 and 19 respectively than in grain from normal parents (Tables 3 and 4). The average oil levels in TopCross grain from TC Blends with pollinators 18 and 19 were 4.1 and 3.6 percentage points greater respectively than in grain of normal parents and 1.9 percentage greater than in grain of P34M55 (Tables 3 and 4).

In 1995, the oil content of TopCross grain produced by the five TC Blends was not significantly different nor were there differences in oil content among the six normal hybrids (Table 1). However, in 1996 there were significant differences in grain oil content among the TC Blends with oil levels ranging from 6.4% to 7.9% (Table 2). There were significant but smaller differences in oil content among the normal hybrids in 1996. In 1997 TopCross grain oil levels also differed significantly, varying by 1.4 and 0.7 percentage points among SuperKernoils with pollinator 18 and 19 respectively (Tables 3, 4). Differences in grain oil content among the normal hybrids in 1997 were not significant. Lysine content and metabolizable energy were higher, whereas starch levels were lower in TopCross grain compared to normal grain. Harvest grain moisture of three of the TC Blends containing pollinator 2 was significantly greater than that their respective normal parents in 1995 but differences were not significant in 1996 (Tables 1 and 2). Grain moistures of one of the five TC Blends containing pollinator 18 and three of the five TC pollinators containing pollinator 19 were significantly higher than their respective normal parents in 1997 (Tables 3 and 4). Stalk lodging and percent of plants barren or with nubbin ears were similar for the TC Blends and normal hybrids (data not shown).

Replicated Plot Study

In 1995 (Table 3) TC Blends averaged 6 Bu/A less than the grain parent checks and 8 Bu/A less than P3394. However, Pfister SK3034-2 and SK2650-2 produced yields comparable or superior to P3394. Three of the TC Blends produced yields comparable to their grain parent checks. Yields of four of the five TC Blends, which varied from 112 to 145 Bu/A, were significantly higher than high oil hybrid CM652HO. In 1996 (data not shown) yields were sharply reduced by late planting (June 17) and severe moisture stress in August during the grainfill period. Grain yields for the test site averaged less than 80 Bu/A. TC Blend grain yields were not significantly different from their respective grain parents. Yields of the high oil hybrid were half those of the TC Blends and grain parent checks. Given the late planting date, considerably beyond that recommended for corn, and late season drought, these results are of limited value for assessing differences in yield potential of normal and high oil corn. Nevertheless, the results suggest that the loss in yield potential associated with late planting and severe drought was no greater for the TC Blends than for grain parent checks. In 1997 (Table 6 and 7) TC Blends averaged 29 Bu/A and 19 Bu/A less than their normal corn counterparts at Hoytville and South Charleston, respectively. Two of the five TC Blends produced yields not significantly different from their respective grain parent checks at Hoytville and four of the TC Blends produced yields not significantly different from their respective grain parents at S. Charleston. At Hoytville three TC Blends produced yields comparable to P3394 and at S. Charleston, four TC Blends produced yields similar to P3394. At Hoytville TC Blend yields were comparable to those of P34M55 whereas at S. Charleston three of the five TC Blends yielded significantly more than P34M55.

The harvest population for the TC Blends was about 9% less than that of the normal hybrids in 1995 (Table 5) but differences in stand were not significant in 1996 (data not shown). Differences in harvest population between TC Blends and grain parent checks in 1997 were not significant at S. Charleston but at Hoytville final stands averaged 12% lower for the TC Blends than for the grain parent checks. These differences in final stand may have contributed to lower the grain yields of certain TC Blends compared to grain parent checks in 1995 and 1996.

Grain moisture levels at harvest averaged 2.3 and 3.2 percentage points higher for TopCross grain in 1995 (Table 5) and 1996 (data not shown), respectively. Grain moisture levels at Hoytville in 1997, where grain drydown was slow, averaged 4.4 percentage points higher for TopCross grain produced by TC Blends than for normal corn grain. These differences were not evident at S. Charleston, where grain drydown was more rapid and most grain moisture levels averaged less than 20% at harvest.

Although planting dates varied considerably for 1995 and 1996, differences in grain composition between the high oil and normal corn were similar both years.TopCross grain produced by the five TC-Blends with pollinator 2 was characterized by a higher oil content than grain of CM652HO (6.9% vs. 5.6% in 1995; 6.9% vs. 5.4% in 1996), the normal parent hybrids, (6.9% vs. 3.8%), and P3394 (6.9% vs. 3.5%). Oil levels for TopCross grain produced by TC Blends with pollinator 19 in 1997 averaged 3.5 and 3.1 percentage points greater than for normal corn, at Hoytville and S. Charleston, respectively (Tables 6 and 7). Starch levels were lower in TopCross grain than than in grain of the high oil single cross hybrids and grain parent checks, but metabolizable energy estimate levels were higher in TopCross grain. Grain protein levels were significantly higher in TopCross grain from some of the TC Blends compared to check hybrids in 1995-1996, but in 1997 no significant differences in protein were evident. Estimates of grain lysine content were higher in TopCross grain from some of the TC Blends than in certain grain parent checks. Lodging and numbers of nubbin ears/barren plants were comparable for the TC Blends and grain parent checks each year (date not shown). Stalk lodging was significantly greater for CM652HO in 1995-1996 and P34M55 in 1997 at both sites than for TC Blends and grain parent checks.

Summary

Field studies conducted from 1995 to 1997 indicate that grain yields of TC Blends averaged 8% to 15% lower than normal hybrids. However test results suggest that certain TC Blends are available with grain yields comparable to normal hybrids. TopCross grain exhibited oil levels about three percentage points higher than grain from the normal check hybrids. TopCross grain moistures at harvest were higher for certain high oil blends compared to their respective normal grain parents but stalk quality was comparable for normal hybrids and TC Blends. High oil single cross hybrids produced yields that were comparable to or significantly less than TC Blends, but their oil levels were about 1.5 percentage points less than TopCross grain. Stalk lodging was consistently greater in the conventional high oil single cross hybrids than in the TC Blends and their grain parent checks.

TC-Blend®, TopCross®, and Optimum® are registered trademarks of Optimum Quality Grains, L.L.C.. SuperKernoil is a trademark of the Pfister Corn Hybrid Company.

Table 1. Comparison of Pfister SuperKernoils (SK) containing Pollinator 2 with normal corn hybrids at five locations in Ohio, 1995.

Table 2. Comparision of Pfister SuperKernoils (SK) containing Pollinator 2 with normal corn hybrids at five locations in Ohio, 1996.

Table 3. Comparison of Pfister SuperKernoils (SK) containing Pollinator 18 with conventional high oil and normal corn hybrids at five Ohio locations, 1997.

Table 4. Comparison of Pfister SuperKernoils containing Pollinator 19 with conventional high oil and normal corn hybrids at seven Ohio locations, 1997.

Table 5. Comparison of Pfister SuperKernoils (SK) containing Pollinator 2 with conventional high oil and normal corn hybrids at Hoytville, Ohio, 1995.

Table 6. Comparison of Pfister SuperKernoils (SK) containing Pollinator 19 with conventional high oil and normal corn hybrids at Hoytville, Ohio, 1997.

Table 7. Comparison of Pfister SuperKernoils (SK) containing Pollinator 19 with conventional high oil and normal corn hybrids at S. Charleston, Ohio, 1997.

Supported in part by an Ohio State University Extension Innovative Grant, the Pfister Corn Company, and the Ohio Corn Marketing Program