|
|
||||||||
Southwest Res. Ext. Cent., 4500 E. Mary, Garden City, KS 67846
Corresponding author (cnorwood{at}gcnet.com)
Received for publication April 17, 2000.
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: D1, first planting date D2, second planting date H1, H2, H3, H4, and H5 are hybrids with 75-, 92-, 98-, 106-, and 110-d maturity, respectively P1, P2, and P3 are 30000, 45000, and 60000 plants ha-1, respectively
| INTRODUCTION |
|---|
|
|
|---|
Corn must be managed properly to tolerate the low precipitation and high temperatures that characterize the central Great Plains, otherwise yield reductions will occur. Important cultural practices include the selection of appropriate hybrids, planting dates, and plant populations. Excluding the last few years, most research on dryland corn in the USA was done in the Corn Belt and Southeast. Planting dates in these areas typically have been in late April or early May (Lauer et al., 1999; Benson, 1990; Swanson and Wilhelm, 1996; Herbeck et al., 1986; Imholte and Carter, 1987). In the western Great Plains, the factors most limiting yields have been rainfall and high temperature. Since dryland corn production began to increase in the early 1990s, the perception has been that it should be planted early, meaning early to mid-April so that it can be pollinated before high midsummer temperatures occur. However, little research has been done to confirm this hypothesis. In north-central and northeast Kansas, Staggenborg et al. (1999) found that planting in early April or early May produced similar yields. In one location, yields were reduced by early planting because ear development occurred after a period of severe drought. Delaying planting to early June reduced yields at all locations. In southwest Kansas, Norwood and Currie (1996) compared early, mid, and late May plantings and concluded that, in most years, dryland corn should be planted in early to mid-May. However, in one year, similar yields were obtained from the early and late-May plantings, and the mid-May planting resulted in the lowest yields. This indicates that stress can occur following any planting date, but dates earlier than 1 May were not evaluated.
Optimum plant populations vary with factors such as moisture stress, hybrid, soil fertility, and yield goal (Benson, 1982). In the central Corn Belt, populations from 54000 to 69000 plants ha-1 are considered optimum, whereas somewhat higher stands are needed in the northern Corn Belt to maximize yield (Benson, 1990). Alessi and Power (1974) found lack of available water during grain formation and filling to be the factor limiting grain production in the Great Plains. The authors reported that optimum populations were 30000 to 40000 plants ha-1 in North Dakota. Results were similar in southern Alberta where a population of 30000 plants ha-1 was recommended (Major et al., 1991). Shorter-season (earlier) hybrids may be necessary to obtain stable yields in areas of low or variable rainfall. In central and eastern Nebraska, Larson and Clegg (1999) found that a full-season hybrid produced a maximum yield at 85000 plants ha-1 if no stress occurred but that populations should be reduced to 45000 to 65000 plants ha-1 under unfavorable environments. However, they suggested that the use of earlier hybrids might improve yield stability because they are not as dependent on yearly weather conditions. In north-central and northeast Kansas, increasing the population from 35000 to 50000 plants ha-1 increased yield by 0.88 Mg ha-1, and increasing the population to 65000 plants ha-1 resulted in an additional 0.25 Mg ha-1 (Staggenborg et al., 1999). A full-season hybrid generally produced more grain than a short-season hybrid when planted early, and growing season length was not a yield-limiting factor. Yields of a short-season hybrid were equal to or greater than yields of a full-season hybrid at later planting dates.
In southwest Kansas, Norwood and Currie (1996) found that the population of a 105-d hybrid planted in early to mid-May should not exceed 45000 plants ha-1. In the driest year of their study, however, a population of 28000 plants ha-1 produced the highest yield. However, no research with different hybrids, populations, and planting dates has been conducted. Because of higher temperatures and lower relative humidities in western Kansas, research from areas further north and east cannot be used. Therefore, the objectives of this study were to evaluate the effects of early vs. later planting, hybrid maturity, and plant population on yield of dryland corn in southwest Kansas.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Corn was planted in a wheatcornfallow rotation, which consists of two crops in 3 yr and a 10- to 11-mo fallow period preceding each crop. Corn was planted with a Buffalo slot planter into the stubble remaining from the preceding wheat crop in mid-April (D1) and early May (D2) of each year. Five corn hybrids differing in relative maturities were planted in 1997 through 1999, and only the three latest hybrids were planted in 1996. Hybrids were Pioneer Brand1 3984, 3860, 3737, 3514, and 3394 and had relative maturities of 75 (H1), 92 (H2), 98 (H3), 106 (H4), and 110 (H5) d, respectively. The three latest hybrids were selected based on their adaptation to the area. The H3 and H4 hybrids have been successfully grown under dryland conditions in the area. The H5 hybrid was probably more suited for limited irrigation, but it was included because it represents the upper end of maturity for dryland. The two earliest hybrids were selected to represent hybrids with earlier relative maturities. Because little research on different hybrids of dryland corn has been conducted in southwest Kansas, selection was based on recommendations from Pioneer Hi-Bred International. The hybrids were planted at a rate of about 87000 seeds ha-1 and hand-thinned after emergence to 30000, 45000, and 60000 plants ha-1 populations. Seed destruction by ground squirrels (Spermophilus spp.) in 1996 resulted in a plant population of 55000 plants ha-1 rather than 60000 plants ha-1 in 1996. The corn was no till in 1996, 1997, and 1998, but tillage was used before planting in 1999 to destroy the ground-squirrel habitat.
Herbicides used each year were atrazine [6-chloro-N-ethyl-N'-(1-methlyethyl)-1,3,5-triazine-2,4-diamine] applied at a rate of 2.2 kg ha-1 in July after wheat harvest followed by 1.1 kg ha-1 atrazine plus 1.5 kg ha-1 dimethenamid {2-chloro-N-[(1-methyl-2-methoxy)ethyl]-N-(2,4-dimethyl-thien-3-yl)-acetamide} in early May for in-crop weed control. The plots received 90 kg N ha-1 before planting each year. The soil P level averaged about 45 kg ha-1, or about the middle of the medium soil test range (Whitney and Murphy, 1969), but 100 kg ha-1 P2O5 was applied at the beginning of the study to eliminate any potential deficiencies.
The experimental design was a split-split plot with four replications. Planting date was the main plot, hybrid was the subplot, and plant population was the sub-subplot. Each sub-subplot was 3 m wide (four 75-cm rows) by 9 m long. Ears from the center two rows were counted and hand-harvested; yield was corrected to 155 g kg-1 moisture; and the yield components ears ha-1, kernel weight (determined from a 100-kernel subsample), kernels ear-1, and kernel weight ear-1 were determined. The corn was harvested as it matured, usually at a moisture content of 150 to 200 g kg-1. A few dropped ears and lodged plants occurred. Although these numbers were not recorded separately, all ears were harvested.
Statistical analysis was by PROC ANOVA with mean separation by Fisher's protected LSD and by the linear forward selection component of PROC REG (SAS Inst., 1998).
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
|
|
The date differences may have been due to cold soil, resulting in slow germination and slow growth early in the growing season, followed by high temperatures and low rainfall during pollination. Colder soils under no-till or high-residue conditions have been noted in other studies (Swanson and Wilhelm, 1996; Herbeck et al., 1986; Griffith et al., 1973; Willis et al., 1957). Colder soils reduced root and shoot dry weights (Kasper et al., 1987), required more growing degree days for the growing point to reach the soil surface (Swan et al., 1987), and modified leaf area (Bollero et al., 1996). Walker (1969) found seedling dry weights to be 20% greater with each 1° increase in soil temperature from 12 to 26°C. Imholte and Carter (1987) found that colder soils under no-till conditions were associated with reduced corn emergence, delayed silking, and increased harvest moisture. Most of these researchers indicated that planting corn into heavy residue should be delayed several days until the soil warmed up.
Most dryland corn in the western Great Plains is planted no till, meaning that the soils are probably colder than tilled soils, particularly in April. No till is highly recommended in western Kansas (Norwood and Currie, 1996, 1997), but no till combined with the early planting date probably contributed to the reduced yields in this study. Soil temperatures averaged 2.8 to 4.0°C lower during the last 2 wk of April vs. the first 2 wk in May (data not shown). The 4-yr average was 13.9°C for late April vs. 17.4°C in early May. The local weather records do not include a compilation of daily average soil temperatures; however, the 30-yr average temperature for the month of April is 13.1°C, meaning the late April soil temperatures were slightly below average.
The most visible stress during pollination occurred in 1998, but the highest yield reduction from early planting occurred in 1997. The H5 hybrid was affected the most in 1997, particularly at P3. Late June through early July was the pollination period for most of the hybrids. Although June 1997 rainfall was 68 mm above average, only 35 mm fell between 26 June and 6 August, and the largest amount during this period was 10 mm on 6 July and 14 July. Maximum temperatures were near normal (30°C) during this period and combined with the low rainfall to reduce yields. Under the conditions of this study, there was no advantage to early planting. Although the number of days to maturity were not measured in this study, the difference among the number of days to maturity was less than the number of days between planting dates. Dryland corn yields are affected by climatic conditions during critical growth stages, particularly pollination. Visible stress, as evidenced by leaf rolling, occurred by late June in all years except 1996 at about the time when the earliest hybrids from D1 were beginning pollination. Therefore, in this study, the reduction in yield of D1 was probably due to colder soil during germination and early growth followed by stress during pollination and grain fill. Although temperatures were usually as high or higher during pollination and grain fill following D2 as they were following D1, germination and early growth also occurred during warmer temperatures.
This study did not necessarily prove that early planting will always be detrimental because the probability exists that unfavorable climatic conditions can follow any planting date. To reduce risk, farmers should plant on more than one date but, on average, planting as early as mid-April will probably reduce yield.
The climatic conditions during the study generally favored the higher yield potential of the later maturing hybrids. The H1 hybrid simply did not have the potential to produce an adequate yield compared with the other hybrids. At the opposite end of the spectrum, H5 yielded more than the other hybrids in nearly all statistical comparisons, except for D1 in 1997 or compared with H2 and H4 for D2 in 1997 when there was no difference.
Favorable climatic conditions usually resulted in the higher populations producing more grain. However, during the stressful year of 1997, average yield for D1 decreased from 3.39 Mg ha-1 at P1 to 2.60 Mg ha-1 at P3 while 45000 plants ha-1 (P2) and P3 produced the most yield for D2. Yield of H5 was less at P3 for both D1 and D2, but the largest reduction, 2.68 Mg ha-1 (66.3%), occurred for D1 when the population was increased from P1 to P3. Higher populations decreased yield of all hybrids planted on D1 except H3 in 1997. The largest positive responses to population occurred in 1996, a year of good rainfall distribution. The average increases in yield between P3 and P1 in 1996 were 3.07 Mg ha-1 (54.5%) on D1 and 3.34 Mg ha-1 (51.2%) on D2. The 1997 through 1999 average indicates that earlier hybrids responded more to increases in population than later hybrids. Average yield increases from P1 to P3 were H1, 36.7; H2, 24.2; H3, 28.5; H4, 13.6; and H5, 3.5%. The increase in yield with increase in population was not linear, however. The average increase in yield from P1 to P2 was 13.5%, whereas the average increase in yield from P2 to P3 was 4.3%. Other than in 1997, increased population had no disadvantage. However, precipitation was above average during the study period, and the yields from 1997 indicate yield decreases can occur if populations are too high. Risk-averse farmers probably would choose lower populations.
Yield Components
Figures 1 and 2 compare ears ha-1 and kernel weight ear-1 for the date x hybrid and hybrid x population interactions, respectively, and Fig. 3 and 4 compare kernels ear-1 and kernel weight for those interactions. For brevity, the date x population interaction is omitted because it does not add much to this discussion. No significant three-way interactions occurred. Grain yield is the product of ears ha-1, kernels ear-1, and kernel weight (yield also can be expressed as the product of kernels ha-1 and kernel weight). Kernel weight ear-1 is determined by the number of kernels ear-1 and kernel weight. With the exception of 1997 (discussed below), there were very few barren plants. Thus, the number of ears ha-1 (Fig. 1) was determined mostly by the plant population. The largest difference in ears ha-1 between hybrids occurred in 1997 when H1 at P3 produced nearly 60000 ears ha-1 and H5 at P3 produced only 38000 ears ha-1. Thus, about 40% of the plants were barren in 1997 for H5 at P3. Only H5 in 1997 did not produce more ears as the population increased from P2 to P3. However, kernel weight ear-1 (Fig 1) typically declined as population increased. Even in 1996, a year of excellent rainfall distribution, kernel weight ear-1 of H4 and H5 declined slightly from P1 to P3. Kernel weight ear-1 of H3 did increase slightly as population increased from P1 to P2 in 1996, but the difference was not statistically significant.
|
|
|
|
The number of kernels ear-1 decreased with increased population for most of the hybrids, whereas kernel weight either declined slightly or did not change (Fig. 3). The number of kernels ear-1 for H3 did increase slightly in 1996 as population increased. The H5 hybrid almost always had the most kernels ear-1, whereas H1 had the fewest. The second planting date usually resulted in the most kernels ear-1 (Fig. 4), and the largest difference between planting dates occurred in 1997. Only slight differences in kernel weight occurred between planting dates in 1996. Kernels were heavier for D2 in 1997, but differences between dates were not consistent in 1998 and 1999. It is interesting to note that the 92-d hybrid (H2) produced heavier kernels than did the 98-d hybrid (H3) for both planting dates in 1997, 1998, and 1999 (Fig 4). Yields of H2 and H3 did not differ significantly in 1997 (Table 2), but in 1998 and 1999, H3 yielded more than H2 because of a greater number of kernels ear-1.
The reasons for the differences in the yield components among years and planting dates can be partially explained by climatic conditions at critical growth stages. Ritchie et al. (1997) stated that potential ears are formed at V5 (five-leaf stage), potential kernels at V12, and potential kernels per row by V17 (about 1 wk before silking). The years 1997 and 1998 can be used as examples. Rainfall in June 1997 was above average, whereas rainfall in July 1997 was well below average. The June rainfall was adequate for potential ear size, but in 1997, the lack of July rainfall restricted ear development in that the number of ears ha-1 (Fig. 2), kernels ear-1 (Fig. 4), and kernel weight (Fig. 4) of all hybrids was reduced with D1. This contrasts with 1998 when June rainfall was low, and July rainfall was well above average. Apparently, even the low rainfall in June 1998 was adequate for potential ear size, and the above average July rainfall allowed the ears to develop. In 1998, the number of ears ha-1 (Fig. 2) was slightly higher for D1 than D2, but this was more than compensated for by more kernels ear-1 following D2 (Fig. 4).
Forward selection regression was used to determine the relative effects of yield components on yield (Table 3). The R2 values are measures of the effects of each component on the variation in yield. Kernels ear-1 accounted for 58 to 64% of the variation in yield in 1997 through 1999, and ears ha-1 accounted for 26 to 34%. Kernel weight caused only about 5% of the variation. Differences in kernel weight among hybrids, years, and dates did occur (Fig. 4), but differences in kernel weight were not nearly as important as ears ha-1 and kernels ear-1. In 1996, ears ha-1 caused almost three times more variation than did kernels ear-1. The two earliest hybrids produced fewer kernels ear-1 than the later hybrids, and their absence in 1996 resulted in kernels ear-1 having less overall influence on yield. The effects of ears ha-1 and kernels ear-1 differed among individual hybrids. Ears ha-1 accounted for more of the variations in yield of H1 and H4 during the last 3 yr, whereas kernels ear-1 accounted for more variation in the yields of the remaining hybrids. Kernel weight accounted for only 2 to 6% of the yield variation and was nonsignificant for H5.
|
| SUMMARY |
|---|
|
|
|---|
| NOTES |
|---|
|
|
|---|
1 Mention of a trade name does not imply endorsement by Kansas State University over comparable products. ![]()
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Y. Kamara, F. Ekeleme, D. Chikoye, and L. O. Omoigui Planting Date and Cultivar Effects on Grain Yield in Dryland Corn Production Agron. J., January 8, 2009; 101(1): 91 - 98. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. A. Norwood Planting Date, Hybrid Maturity, and Plant Population Effects on Soil Water Depletion, Water Use, and Yield of Dryland Corn Agron. J., September 1, 2001; 93(5): 1034 - 1042. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||