| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Hortic. and Crop Sci., The Ohio State Univ., Columbus, OH 43210-1086
b The Ohio State Univ. Ext., Columbus, OH 43210-1086
* Corresponding author (thomison.1{at}osu.edu)
Received for publication January 12, 2001.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
Abbreviations: GDD, growing degree day • HO, high oil • HOC, high-oil corn • NWBRF, Northwest Branch Research Farm • WBRF, Western Branch Research Farm
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
According to the U.S. Grains Council (1999), HOC production in the USA has increased from <20000 ha in 1992 to >400000 ha in 1999. The TopCross1 grain production system licensed by DuPont Specialty Grains (Des Moines, IA) is rapidly gaining popularity as the preferred method of producing HOC (U.S. Grains Council, 1999). Commercial high oil (HO) single-cross hybrids have not been widely used by growers because their grain yield potential is lower than normal dent hybrids (Watson and Freeman, 1975; Lambert, 1994). The TopCross system may minimize the yield disadvantage associated with conventional HOC hybrids while enhancing grain nutrient composition (Edge, 1997; Lambert et al., 1998). With the TopCross system, HOC has been reported to yield as well as normal corn (Cromwell, 2000).
The TopCross HO grain production system involves planting a blend (TC Blend) of two types of corn (Edge, 1997). 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 in the blend, is designated as the special pollinator. The TC Blend grain parent is a male sterile (produces no viable pollen) version of an elite hybrid. The TC Blend pollinator is a special line, available from DuPont Specialty Grains and licensed to seed companies. According to DuPont Specialty Grains, pollinators are either synthetics, pseudo hybrids, or open-pollinated populations (Edge, 1997). The primary function of pollinators is to provide pollen to male sterile grain parents; however, they contribute little to grain yield. The pollen shed from these pollinator plants contain genes that cause production of kernels with larger-than-average embryos (Lambert et al., 1998). Because most of the oil and essential amino acids are in the embryo, the increased embryo size of HOC results in greater oil content and enhances the protein quality of the grain (Cromwell, 2000).
Although HOC may be a profitable alternative to normal (low oil) corn, production costs associated with TopCross corn are greater. The TC Blend seed is more expensive than normal hybrid seed, and growers may need to use higher seeding rates to compensate for the lower grain yields of pollinator plants in TC Blends (Thomison et al., 1997). According to DuPont Specialty Grains, the TopCross system may be more vulnerable to environmental stress at pollination due to the limited number of pollinator plants in TC Blends (Thomison et al., 1997).
Aside from seed company literature, little information is available on the agronomic performance of TC Blends used in the TopCross HOC production system. Compared with conventional hybrids, evaluating TC Blends is difficult due to isolation requirements. Most university performance trials of TC Blends have not included conventional (low oil) counterparts. This has precluded direct comparisons of yield potential and other economically important agronomic traits, such as grain drydown, test weight, and stalk lodging. The agronomic performance of TC Blends relative to conventional (low oil) corn hybrids will be an important consideration for growers assessing the additional costs associated with the production of TC Blends.
The primary objective of this study was to compare the agronomic performance of TC Blends used in TopCross production systems with that of conventional hybrid counterparts used in commodity grain production. We especially wanted to assess effects of the limited number of pollinators in the TopCross system on successful pollination and kernel set. Because there have been improvements in the yield potential of pollinators used in TC Blends during the past 5 yr, this study of the TopCross system included several pollinator types that have been used commercially.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
Experiment 1
Plots were established at The Ohio State University Ohio Agricultural Research and Development Center (OSU-OARDC) Northwest Branch Research Farm (NWBRF) near Hoytville in northwest Ohio in 1995 and 1996. Five TC Blends (Pfister brand SuperKernoils SK3034-2, SK3001-2, SKX577-2, SKX592-2, and SK2650-2) were compared with low-oil conventional hybrid counterparts (Pfister brand hybrids 3034, 3001, X577, X592, and 2650; El Paso, IL). The relative hybrid maturity and growing degree day (GDD) ratings (from VE to R6; Ritchie et al., 1989) of the conventional hybrids (hereafter referred to as check hybrids) ranged from 108 to 110 d and 1483 to 1500 GDDs, respectively. These relative maturity ratings are specific to central Corn Belt growing conditions.
Experiment 2
Five TC Blends containing Pfister pollinator 19 (SK3049-19, SK2650-19, SK3001-19, SK2680-19, and SK2652-19) and their respective check hybrids (Pfister brand hybrids 3049, 2650, 3001, 2680, and 2652) were grown at NWBRF and OSU-OARDC Western Branch Research Farm (WBRF) near South Charleston in southwest Ohio in 1997. The relative maturity and GDD ratings (from VE to R6; Ritchie et al., 1989) of the check hybrids ranged from 108 to 110 d and 1450 to 1540 GDDs, respectively.
Experiment 3
Three TC Blends and their respective check hybrids were planted at NWBRF and WBRF in 1998 and 1999. Each TC Blend was associated with a different pollinator. The TC Blends evaluated were DEKALB DK595TC, Pioneer Brand 34K79, Pfister SK3049-19 (in 1998), and Pfister SK2652-19 (in 1999); the check hybrids were DEKALB DK595, Pioneer Brand 34K77, Pfister 3049 (in 1998), and Pfister 2652 (in 1999). The relative maturity and GDD ratings (from VE to R6; Ritchie et al., 1989) of the check hybrids ranged from 108 to 110 d and 1450 to 1540 GDDs, respectively.
On-Farm Study 1
The same TC Blends and check hybrids used in Exp. 1 were established in non replicated strip plots at 10 on-farm sites near South Charleston, Washington Courthouse, Columbus, Hebron, Van Wert, and Hoytville in 1995 and 1996.
On-Farm Study 2
The same TC Blends and check hybrids used in Exp. 2 were established in non replicated strip plots at seven on-farm sites near South Charleston, Columbus, Bellefontaine, Hebron, London, Wooster, and Hoytville in 1997. Each on-farm site was in a different Ohio county.
On-Farm Study 3
Five TC Blends containing Pfister pollinator 18 (SK3034-18, SK2020-18, SK2025-18, SKX777-18, and SK2320-18) and their respective check hybrids (Pfister brand hybrids 3034, 2020, 2025, X777, and 2320) were established in nonreplicated strip plots at five on-farm sites near Columbus, Hebron, London, Wooster, and Hoytville in 1997. The relative maturity and GDD ratings (from VE to R6; Ritchie et al., 1989) of the check hybrids ranged from 105 to 107 d and 1350 to 1440 GDDs, respectively.
In each experiment and on-farm study, treatments were arranged as a split-plot arrangement of treatments in a randomized complete block design. Type of corn, HO (TC Blend) vs. low oil (check hybrid), was assigned to the whole plot, and grain parent was assigned to the subplot. Grain parent refers to the genetic background common to the fertile check hybrid and the male sterile TC Blend grain parent. In Exp. 1 through 3, three replications were used, except in 1995, when five replications were used in Exp. 1. To determine if variation in TC Blend and check-hybrid performance was significantly different for the nonreplicated on-farm strip tests, data from the strip tests were combined in each of the on-farm studies, with locations treated as blocks (replications), and analyzed as a randomized complete block split plot (10 replicated blocks for On-Farm Study 1, seven for On-Farm Study 2, and five for On-Farm Study 3).
The unique nature of the TopCross grain production system makes an effective evaluation of TC Blend performance in conventional small plots difficult and questionable. Due to the limited number of pollinator plants in a TC Blend and the resulting reduction in pollen shed at anthesis, as well as the potential for xenia effects (effect of pollen from neighboring normal corn), we conducted these evaluations of agronomic performance using field-scale strip plots that would better approximate and simulate a comparison of HOC and conventional corn performance in grower fields. In Exp. 1 through 3, each TC Blend and check-hybrid plot included four rows spaced 0.76 m apart and 61 m in length. In the on-farm evaluations, each TC Blend and check hybrid was planted in a nonreplicated strip plot at least four rows wide (0.76-m row spacings) and 61 m in length. All plots were planted at a density of at least 76000 seeds ha-1.
To minimize contamination of TC Blends by pollen from check hybrids, the following testing protocol was used for comparing TC Blends and their fertile grain-parent counterparts. At each location, a 61-m buffer was planted with TC Blend seed to separate the TC Blend plots from the neighboring check hybrids. This minimized foreign pollen contamination of the TC Blends. Borders (6.1–15.2 m) on other sides of the isolation field were also planted with TC Blend seed to minimize edge-row effects and ensure adequate pollen shed. At all locations, the only nearby foreign pollen source was that of the grain-parent check hybrids. Figure 1 shows the arrangement of the TC Blend and check-hybrid strip plots in relation to the buffer and borders for the on-farm test sites; similar buffers were used to separate check hybrids and TC Blends in Exp. 1 to 3. This method for evaluating TC Blends and check hybrids is similar to that used by various seed companies (Gaspar, 2000; Thomison et al., 1997).
|
Nutrient, insect, and weed management strategies appropriate for minimizing crop stress were followed at each location. Table 1 indicates soil types and planting dates associated with each evaluation. The previous crop at each location was soybean [Glycine max (L.) Merr.], except at NWBRF in 1998, where the previous crop was wheat (Triticum aestivum L. em Thell.).
|
For Exp. 1 to 3, data for each trait were subjected to analysis of variance at each location. Data were combined over locations each year for analysis in Exp. 2 and 3. In each experiment and on-farm study, treatment mean comparisons were made using least significant differences at the 0.05 probability level (LSD 0.05).
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
|
|
|
Experiment 2
Averaged across two locations (WBRF and NWBRF), yield, grain moisture, test weight, plant population, lodging, 300-kernel weight, and kernels ear-1 of TC Blends containing pollinator 19 were significantly different from check hybrids (Table 3). These traits were also affected by grain parent. There were significant environment x type interactions for several of the agronomic traits, including yield, which were attributed primarily to differences in magnitude vs. rank changes at the two sites (Table 3). Average yield for the five TC Blends was 10 and 14% lower than that of the check hybrids at NWBRF and WBRF, respectively (Table 5). Grain moisture levels at NWBRF in 1997, where grain drying was slow, averaged 44 g kg-1 higher for HO grain produced by TC Blends than for check-hybrid grain. These differences were not evident at WBRF where grain drying was more rapid and most grain moisture levels averaged <200 g kg-1 at harvest. Averaged across locations, test weights averaged 11.5% lower in TC Blends than in check hybrids. Differences in harvest populations between TC Blends and grain parent checks in 1997 were not significant at WBRF, but final stands averaged 12% lower for the TC Blends than for the grain parent checks at NWBRF (Table 5). Averaged across locations, levels of stalk lodging (data not shown) were negligible but slightly higher in TC Blends than in check hybrids (4 vs. 2%). Averaged across locations, the TC Blends averaged more kernels ear-1 than the check hybrids but lower kernel weights.
|
|
|
|
|
On-Farm Test 3
Yields of TC Blends containing pollinator 18, averaged across five on-farm test sites in 1997, were not significantly different from the check hybrids (Tables 8 and 9). In four of the five comparisons, the check hybrid yielded more than the TC Blend, but the average difference failed to achieve significance despite a yield difference of 9%. The check hybrid averaged higher test weights than the TC Blends. Other agronomic characteristics were not significantly different between TC Blends and check hybrids. Most of these agronomic characteristics differed significantly among the grain parents, but there were no significant corn type x grain parent interactions (Table 8).
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
A 1995 University of Wisconsin evaluation comparing the agronomic performance of three TC Blends and their normal counterparts indicated that yields averaged 10% less and grain moisture contents 17 g kg-1 more in TC Blends (Lauer, 1995). Averaged across two Minnesota locations in 1997, TC Blends were 12, 10, and 7% less than conventional hybrids in early, mid-, and late-maturity groups, respectively (D.R. Hicks, personal communication, 1998). Test weight and kernel moisture content of grain from TC Blends were similar to the conventional hybrids.
Several factors may account for the lower yields of TC Blends compared with their respective grain parent checks. Because the pollinator in a TC Blend does not contribute as much grain to crop yield as male sterile or fertile grain parents (Edge, 1997), given comparable final stands, the yield of a TC Blend may be lower than a normal counterpart due to a lower number of grain parent plants. Moreover, TC Blend pollinators also use resources—light, water, and nutrients—at the expense of the grain parents. Because the TC Blend pollinator and grain parent plants are similar in size, light intercepted by TC Blend pollinators may be lost in some proportion to the numbers of pollinators present (Rimmington, 1984), and thereby impact photosynthesis and potential yield of the TC Blend grain parents.
The lower plant populations of TC Blends compared with check hybrids may have contributed to lower TC Blend yields. Although not always significantly different, plant populations of TC Blends were lower than check hybrids in most comparisons (Tables 4, 5, 7, and 9). The lower stands of TC Blends may be due to reduced emergence and/or increased mortality of pollinator plants, which may have less seedling vigor than hybrid checks (Edge, 1997). According to seed labels, the TC Blends used in this study contained about 9% pollinator seed. Estimates of pollen-shedding plants in TC Blends ranged from 3 to 7% in our experiments (data not shown). This suggests that some of the stand loss in TC Blends compared with check hybrids was due to a reduced number of pollinators.
According to the seeding rate guidelines of DuPont Specialty Grains and various seed companies (Thomison et al., 1997), TC Blends should be planted at 4940 kernels ha-1 higher density than typical up to a planting rate of 74100 kernels ha-1, and above that rate, the same rate as conventional hybrid seed should be used. These higher seeding rates are recommended to compensate for the lower grain yields of pollinator plants in TC Blends. In this study, planting rates of at least 74500 seeds ha-1 were selected, which were optimal for the check hybrids as well as for the TC Blends based on seed company guidelines, and we thereby avoided the need to use different seeding rates for the TC Blend and check hybrid. Results of these experiments suggest that even higher seeding rates may be needed to compensate for the potential of lower TC Blend stands. However, seed company agronomists discourage using seeding rates in excess of those used to optimize conventional corn yields because the greater competition among plants associated with high plant density may adversely affect pollinator performance, and thereby jeopardize adequate pollen production (Thomison et al., 1997). Because TC Blend seed is more expensive than conventional hybrid seed, increasing seeding rate also reduces the profitability of HOC production. Gaspar (2000) recently evaluated plant population responses for TC Blends and their normal counterpart hybrids within a range of 44460 to 88920 plants ha-1 and concluded that the optimum plant populations for yield were generally similar for TC Blends and their normal counterparts.
Differences in yield between TC Blends and check hybrids were not consistently associated with lower plant populations. Most differences in population were small and not of the magnitude that would normally cause significant differences in yield, especially across such a wide range of environmental conditions (Ohio State Univ. Ext., 1995).
Another factor that may contribute to the lower yields of HOC compared with conventional corn is the physiological cost of oil synthesis. Oil in the kernel requires 2.25 times the energy input as the same amount of starch. With a fixed amount of energy available from the plant (Penning de Vries et al., 1974; Lambert et al., 1998), the grain yields of HOC would be lower if the HO hybrid could not trap more solar energy than a normal hybrid. Alexander and Lambert (1968) performed an experiment to test for negative association of HO and grain yields in corn and concluded that oil content and caloric yield plant-1 are probably independent in most hybrids, but calorie yield may be limited by an inefficient oil-synthesizing mechanism in some hybrids. In a subsequent study, Lambert et al. (1998) reported that by using elite commercial male sterile hybrids and an HO pollinator, the TopCross approach increases kernel embryo size of the male sterile hybrid, thus elevating oil content with little effect on grain yield. However, that study did not use commercial TC Blends but rather two-row plots of detasselled male sterile hybrids bordered by an HO pollinator (Lambert et al., 1998); yields were determined on two rows of male sterile hybrids that did not contain pollinators.
Use of male sterile grain parents in TC Blends may enhance agronomic performance and offset some of the lower yield potential of pollinators. Cytoplasmic male sterile maize hybrids have produced higher yields than their normal fertile counterparts, which is assumed to result from no pollen production (Duvick, 1958; Russell and Marquez-Sanchez, 1966).
These evaluations provide little evidence that TC Blends are more vulnerable to hotter and drier conditions during pollination. The yield disadvantage associated with TC Blends relative to check hybrids was not accentuated by drought conditions associated with certain test sites (Table 2). Previous efforts to introduce HOC production failed partly because the HO hybrids available were less tolerant of drought than conventional hybrids (Lambert, 1994). Yield differences between TC Blends and check hybrids at locations with below-normal rainfall (WBRF in 1998–1999; NWBRF in 1997 and 1999) were comparable to differences at locations with normal or above-normal rainfall (WBRF in 1997 and NWBRF in 1998) (Table 2). Asynchronous pollen shed and silking caused by drought often results in barrenness or incomplete kernel set (Bassetti and Westgate, 1993). In our evaluation, numbers of barren plants, averaged across corn types, were generally <3% in each experiment and were similar between TC Blends and check hybrids. There was also little or no evidence that kernel set in TC Blends was limited by inadequate pollen availability due to drought or a limited number of pollinators. Measurements of ear yield components indicated no significant differences in kernels ear-1 between TC Blends and check hybrids in most comparisons; however, TC Blends were usually associated with more kernels ear-1. Significant type and type x grain parent effects for incomplete kernel set were only observed in Exp. 1 in 1995 at NWBRF (Table 4) when one TC Blend exhibited higher levels of incomplete kernel set than its check hybrid (10 vs. 0%). However, temperatures were near normal, and precipitation was not markedly below normal during pollination at this site in 1995 (Table 2). Kernel weights of TC Blends and check hybrids were generally similar, but test weights of TC Blends averaged significantly less than check hybrids in most comparisons.
Major yield losses associated with poor pollination and reduced kernel set have been observed in TC Blends grown in some farmer fields (Thomison et al., 1997). In Ohio, Thomison et al. (1997) attributed lower yields in TC Blends to silk clipping and root lodging caused by western corn rootworm (Diabrotica virgifera Le Conte); silk clipping and root lodging were much greater in TopCross corn fields following corn than in fields following soybean or wheat. Similar observations relating yield losses in TC Blends to crop rotation have been reported elsewhere in the Corn Belt and have led to recommendations that TC Blends not be planted after corn (Gaspar, 2000; Thomison et al., 1997). In our evaluations, we planted TC Blends and check hybrids after soybean or wheat, and thereby may have avoided this problem. Seed company agronomists have also reported that asynchronous pollen shedding and silking in certain TC Blends have reduced kernel set and yield in western Corn Belt states.
The TC Blends averaged higher grain moisture than the check hybrids in nearly all comparisons, but differences were usually not significant. Grain moisture contents for TC Blends were 23 g kg-1 more than for check hybrids in comparisons where differences were significant. The higher moisture content of HO grain compared with that of conventional corn has been observed in previous studies (Lauer, 1995; Misevic et al., 1988). Misevic et al. (1988) proposed that the slower drying of HOC may be related to a larger embryo and flintier endosperm of HO kernels. The slightly later date of anthesis associated with TC Blends compared with their respective check hybrids may also affect grain drying. In TC Blends, pollen shed by pollinator plants generally occurs 1 to 3 d after the male sterile grain parent plants initiate silking, whereas in the check hybrids, pollen shed precedes silking by 1 to 3 d (Thomison, unpublished, 2000). This delay in pollen shed by TC Blend pollinators may result in later kernel development, slower kernel maturation, and higher grain moisture at harvest in a TC Blend male sterile grain parent compared with its check hybrid. However, the delay may promote better pollination because there is greater silk emergence when pollen shed begins (Gaspar, 2000).
Stalk lodging was comparable for TC Blends and check hybrids in nearly all comparisons (data not shown). Lodging has been a problem with HOC hybrids in the past (Lambert, 1994; Alexander and Seif, 1963), and even recently introduced single-cross HOC hybrids have exhibited more stalk quality problems than conventional hybrids (Thomison et al., 1997).
|
SUMMARY |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
1 TopCross and TC Blend are registered trademarks of DuPont Specialty Grains, Des Moines, IA. Trade names are used in this publication solely for the purpose of providing specific information. Mention of a trade name, proprietary product, or specific equipment does not constitute a guarantee or warranty by the Ohio State University and does not imply approval of the named product to the exclusion of other products that may be suitable. 
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. F. Mangen, P. R. Thomison, and S. D. Strachan Early-Season Defoliation Effects on TopCross High-Oil Corn Production Agron. J., April 27, 2005; 97(3): 823 - 831. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. R. Thomison, A. B. Geyer, L. D. Lotz, H. J. Siegrist, and T. L. Dobbels TopCross High Oil Corn Production: Select Grain Quality Attributes Agron. J., January 1, 2003; 95(1): 147 - 154. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Plant Registrations | Soil Science Society of America Journal | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
