| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Univ. of Missouri, 108 W. North Main, Richmond, MO 64085 USA
b Dep. of Agronomy and Ctr. for Sustainable Agricultural Systems, 225 Keim Hall, Univ. of Nebraska, Lincoln, NE 68583-0949 USA
lesoingg{at}ext.missouri.edu
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMaterials and methodsResults and discussionSummaryREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionSummaryREFERENCES |
|---|
Water stress, light quality, and shading are among the factors that affect crop yields and yield components at different growth stages. Sixty-three percent shading of soybean plants caused abscission of half the pods (Mann and Jaworski, 1970). Schou et al. (1978) showed that shading soybean plants during reproductive growth influenced seed yield, but seed weight was not changed. Egli and Yu (1991) found that shading from growth stages R1 to R5 in soybean reduced seed yields and seed number, but did not affect seed size. These results are relevant to border soybean rows in strip intercropping.
Effects of water stress on yield and yield components in corn have been studied extensively. Eck (1986) found that water deficits during vegetative growth reduced corn kernel number, but had little effect on weight per kernel. Kernel numbers were not affected by water deficits during grain filling unless severe deficits were imposed early in the period. Harder et al. (1982) also found kernel number to be influenced by early stress, with little influence on seed weight.
In yield component studies in intercropping, Francis et al. (1978) found no significant differences in ear diameter, ear length, row number, 100-seed weight, and harvest index of corn between monoculture and intercropping with bush or climbing bean (Phaseolus vulgaris L.). Research by Willey and Osiru (1972) showed corn to have a higher competitive ability at higher populations when intercropped with bean, probably due to shading effects on bean. In the bean crop, the main component associated with yield decrease was lower pod number per plant, with seed weight not affected. In a four-row corn–soybean strip-cropping system, Wright (1981) found soybean 100-seed weights greater than the same crop in monoculture; corn 100-kernel weights were similar for both cropping systems.
In grain sorghum–soybean intercropping, sorghum seed weight was lower in monoculture than in intercropping (Elmore and Jackobs, 1984; Pavlish, 1989; Wahua and Miller, 1978). Pavlish (1989) found an increase in soybean seed weight; other researchers found no difference in seed weight between monoculture and intercropped systems. Carter (1984) and Pavlish (1989) found a high positive correlation between yield and seed number in both grain sorghum and soybean in intercropping patterns. In grain sorghum, Saeed et al. (1986) found that the number of seeds per unit area was the component most closely related to grain yield.
With development of improved corn and grain sorghum hybrids, new soybean cultivars, new planting equipment, compatible herbicides, and other production practices that can reduce soil erosion and use resources more efficiently, cereal–legume strip-intercropping systems provide an alternative spatial pattern for future systems. If yield components and the spatial cropping patterns that influence their yield contributions can be identified, systems can be designed to increase potential productivity. Objectives of this study were to (i) compare the effects of row position (border rows vs. inside rows) on yields and yield components of corn and soybean in a strip-intercropping system in contrasting environments, (ii) compare the effects of row position on yields and yield components of grain sorghum and soybean in a strip-intercropping system in the same environments and (iii) use yield component data to determine the relative timing of competition for growth resources in strip intercropping compared with monoculture and suggest how this information can contribute to more productive system design.
|
Materials and methods |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionSummaryREFERENCES |
|---|
Corn and soybean were no-till planted in a north–south orientation in alternating 6.1-m-wide strips (8 rows, 0.76 cm between rows), 46 m in length. Experimental units were single strips of 280 m2; these were subsampled for yields and yield components. `Pioneer 3377', a full-season corn hybrid for this zone (114 d), was planted at the rate of 66250 seeds ha-1 at the irrigated site, and the same hybrid at 46500, 44000, and 46250 seeds ha-1 at the rainfed site in 1988, 1989, and 1990, respectively, based on available soil moisture. Previous crop for all corn strips was soybean. Between the corn strips, soybean cultivars were planted in strips of eight 0.76-cm rows at the rate of 475000 seeds ha-1. Four cultivars were included: `BSR 101' (Maturity Group I), `Hack' (Maturity Group II `Hobbit 87' (Maturity Group III), and `Ripley' (Maturity Group IV). Soybean cultivars were reallocated to strips each year, and replicated three times in a randomized complete block design, with a single strip per replication. Previous crop for soybean strips was corn in 1988 and grain sorghum in 1989 and 1990.
Nitrogen fertilizer rates for corn were determined by previous soybean yields. Granular NH4NO3 (34–0–0 N–P–K) was broadcast prior to planting. Nitrogen rates were 153 and 125 (1988), 159 and 102 (1989), and 136 and 91 kg N ha-1 (1990) for irrigated and rainfed corn, respectively. No fertilizer was applied to soybean. Corn planting dates were between 10 May and 23 May, depending on rainfall and conditions appropriate in each year. Soybean planting dates were between 18 May and 1 June.
Grain sorghum–soybean strips were no-till planted in a north–south orientation adjacent to the corn–soybean experiments with the same plot size and planting methods. Grain sorghum hybrid `NC+174', a full-season hybrid (74 d to half bloom), was planted at the rate of 215000 seeds ha-1 at both sites. Sorghum always followed soybean strips. Soybean cultivars and planting details were the same as above. Treatments were replicated four times. Previous crop for soybean was grain sorghum in 1988 and corn in 1989 and 1990.
Nitrogen fertilizer rates for sorghum were determined by previous soybean yields. Granular NH4NO3 (34–0–0 N–P–K) was broadcast prior to planting. Nitrogen rates were 34 kg N ha-1 at both sites in 1988, and 91 and 57 kg N ha-1 for irrigated and rainfed grain sorghum, respectively, in both 1989 and 1990. Planting dates for grain sorghum and soybean were between 18 May and 1 June, depending on rainfall and appropriate planting conditions. We believe that the different planting dates confounded with year effects did not seriously affect results or conclusions.
A severe infestation of grasshoppers (Melanoplus sp.) severely defoliated soybean and grain sorghum plants in 1989 and 1990 at the irrigated site. For grasshopper control, carbaryl (1-naphthyl-N-methylcarbamate) was applied.
In 1988, a 4.56-m section of each corn and grain sorghum row was hand-harvested in each strip to determine individual row grain yield and yield components. With four different cultivars in adjacent soybean strips and three replications, this provided 12 corn strip replications for the current analysis. With four soybean cultivars and four replications of each, this provided 16 grain sorghum replications for the current analysis. In 1989, due to less available labor, only four strips (one randomly selected from each replication) of corn and four of sorghum were hand-harvested. In 1990, all 12 corn strips were harvested, but only four strips of grain sorghum. In 1988, no soybean plants were hand-harvested, and yields were taken from combine harvest of the border and the inside rows of all cultivars. In 1989, all soybean cultivars were hand-harvested for yield in the rainfed experiment, but combine harvest yields are reported for the irrigated site. In 1990, only Hobbit 97 was hand-harvested for both yield and yield components from both the experiments. We recognize that the data set represents different numbers of replications of corn, sorghum, and soybean that are confounded with year effects, but expect minimal influence on our conclusions. We further recognize that soybean yield component data are limited to one cultivar and year, and the results should be interpreted accordingly. Soybean yields are from combine harvest in three site-years, and from hand harvest in three site-years. There was a high correlation between hand and combine harvest; thus, little influence is expected from this confounding.
Yield and yield component data collected included grain yields, plant counts, ear counts (corn), head counts (sorghum), seed counts, and 100-seed weights. Plants per square meter, ears or heads per square meter, seeds per ear or seeds per head, seeds per plant, and seeds per square meter were calculated from the collected data. A logical and minimum data set of yield components in corn, for example, would include seed weight, seeds per ear, ears per plant, and plants per square meter; we added the ears per square meter and seeds per square meter because these are often reported in the literature and because they gave us more information to enrich the discussion of components.
Each strip-intercropping experiment in each site and year was analyzed individually, as nonhomogeneity of variances precluded combining over sites and years. The primary objective of this study was to determine the effect of row position on the yields and yield components of crops within the strips. Inside rows in a strip simulate monoculture conditions, since the major effects of a strip-intercropping pattern are seen in the first or border rows (Pendleton et al., 1963; Ayisi et al., 1997). Analyses of variance were used to evaluate differences in yields and yield components between the middle (inside) six rows and the outside (border) two rows of strips. Since strips were planted in a north–south orientation, half the border rows of corn had adjacent soybean to the east and half had adjacent soybean to the west. The same was true for grain sorghum strips. Comparisons between borders and inside rows in a fixed pattern are statistically valid. Observations between corn and sorghum experiments or over years were used to inductively suggest when competition for growth resources was most acute in the contrasting systems. Since yield component data were available for only one soybean cultivar in one year, any discussion about soybean components must be limited to the conditions in that year. Soybean yield data are averaged over four cultivars, and whole-system performance is based on three years and two sites in each year. A higher estimate of biological potential for the intercropping system compared with monoculture could be calculated by taking only the highest yielding soybean cultivar in each year. This was not done.
System performance was evaluated by using the inside rows to simulate monoculture in each crop, and assuming that half of a farm's area would be dedicated to large fields of each of the two crops in a system. This was compared with strip intercropping, where six inside rows are combined with two border rows in each strip of each crop, again with half of the total area dedicated to each crop. The land equivalent ratio (sum of the ratios of intercrop yields to monocrop yields of component crops) was one measure used to compare the two systems. Another was the gross returns from the contrasting systems, based on corn price of $100 Mg-1, grain sorghum price of $90 Mg-1, and soybean price of $250 Mg-1. These were average prices in Nebraska during the years of the experiments. Use of gross returns does not reflect the net income, since fertilizer costs are lower in sorghum and absent in soybean. Other complications of calculating net returns would include charging the soybean enterprise with higher irrigation costs than necessary, since this crop does not respond to additional irrigation in the same manner as corn, for example, and the two crops in strips were irrigated together. Thus, we report gross returns as one indicator of system economic performance.
|
Results and discussion |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionSummaryREFERENCES |
|---|
Corn Yields and Yield Components
Rainfed and irrigated strip-intercropping corn yields and yield components for 1988 through 1990 are shown in Table 1
. Corn border rows yielded significantly higher than inside rows each year under both rainfed and irrigated conditions. These results agree with previous research where yield increases in a strip-intercropping system were primarily due to increases in the border rows adjacent to soybean (Pendleton et al., 1963; Wright, 1981; West and Griffith, 1992; Fortin et al., 1994). Yield increases were greater under irrigation (average +26%) than at the rainfed site (average +19%).
|
The lowest border-row corn yield increase (+10%) was in the rainfed 1988 experiment, also the lowest-yielding trial of the six site-years. Most of this difference was explained by higher seed weight (+8%), indicating a successful capture of resources during the grain-filling period. In contrast, the highest border-row corn yield increase was at the rainfed (+30%) and irrigated (+31%) sites in 1989, a year with below-average but well-distributed rainfall and the highest corn yields among years. In this year, seed number was higher at both sites (+20%) and seed size was higher by only 5 to 7%. Results indicate that border-row corn plants developed a relatively larger potential factory during the reproductive period, as shown by increased seed number, and were able to supplement this with higher seed size as well. Over the six site–year combinations, there was also a significant negative correlation between border-row increase in seed number and increase in seed weight. This suggests that the larger the potential for greater seed number that is established during the reproductive period, the less potential for plants to increase seed weight as well during the grain-filling period.
Literature reports on corn suggest that timing of stress influenced both seed number and seed weight (Eck, 1986; Harder et al., 1982). Our results are consistent with these reports. Other studies of intercropped corn have conflicting results. In comparing an intensive corn–bean (P. vulgaris L.) intercropping system with monocultures, Francis et al. (1978) found no differences in corn ear length, ear and cob diameter, row number, shelling percent, and weight per 100 seeds. Willey and Osiru (1972) found higher corn yields per plant in an intensive corn–bean mixture than in pure stands; the greatest corn yield increases were found at higher populations. Although significant at only two of six site years, higher plant populations in our border rows were associated with higher yields. Ghaffarzadeh et al. (1997) found more ears per plant, more kernels per ear, and higher seed weight from corn in border rows next to soybean than in the inside rows of the corn strips. West and Griffith (1992) found higher yields for intercropped corn at higher populations than in solid-seeded rows.
Most researchers attribute higher intercrop yields to better utilization of environmental resources, with greater light interception believed to be the most important factor. Data on light measurements in strip-intercropping systems are limited. Measurements taken by P. Boehner (personal communication) in our experiments showed greater light interception by corn border rows compared with inside rows. Experiments by Kasperbauer et al. (1984) and Karlen and Kasperbauer (1989) suggest that light quality (far red:red light ratios) may influence root and shoot development. Rows in a north–south orientation received relatively more far-red light and probably developed more shoots than roots, compared with rows in an east–west direction. Corn yields were higher under nonstress conditions in a north–south row orientation, which may partially explain our greater yield response under irrigation in the north–south corn–soybean strip intercropping.
Grain Sorghum Yields and Yield Components
Rainfed and irrigated strip-intercropped grain sorghum yields and yield components for 1988 through 1990 are shown in Table 2 . Grain sorghum border row yields were significantly higher than inside rows in five of six site–year combinations. Border-row yield increase averaged 23% in rainfed and 25% in irrigated experiments. These results are in contrast to the corn–soybean strip intercropping, where corn border-row increases with irrigation were consistently higher than at the rainfed site. Grain sorghum border rows may have a lower relative potential yield advantage, because of a smaller height difference compared with adjacent soybean rows, less additional light interception, less differential response to irrigation, and/or less competitive advantage over soybean in the root zone.
|
Soybean Yields and Yield Components
Rainfed strip-intercropped soybean yields (1988–1990) and yield components (1990 only) for the corn–soybean system are presented in Table 3
. There was no difference between border and inside rows in 1988, with low yields probably due to lack of rainfall at the right times at the rainfed site and due to grasshopper infestation at the irrigated site. In 1989 and 1990, soybean border-row yields were significantly reduced compared with the inside rows at both rainfed (average -22%) and irrigated (average -28%) sites. This agrees with results of West and Griffith (1992), where soybean border row yields next to corn were reduced by 27%.
|
Rainfed and irrigated strip-intercropped soybean yields (1988 to 1990) and yield components (1990 only) for the grain sorghum–soybean system are presented in Table 4 . Soybean yield differences between row positions were less than in corn–soybean intercropping strips, probably due to the lesser height difference between the two adjacent crop species in this system. Significant border-row yield reductions were seen in only two of six site-years. On average over the three years, soybean yields were reduced 5% in border rows in rainfed conditions, and were the same between row positions at the irrigated site. This suggests that water was the resource for which competition occurred between these two crops of similar plant height. Pavlish (1989) found soybean yield increases in most environments in grain sorghum–soybean intercropping systems compared with monoculture. She concluded that these benefits probably occurred because of minimal interference in light interception and a complementary use of other growth resources. In the experiment by Wahua and Miller (1978), soybean yields were reduced 75% in the tall sorghum intercropping system, but only 17% when intercropped with a semidwarf sorghum cultivar; however, they did not distinguish between competition for light and for other resources.
|
Total System Evaluation
System performance was evaluated in three ways: crop yields, land equivalent ratios, and gross returns per hectare, as shown in Table 5
. In the yield differences reported here, none is significant. At the rainfed site, corn yields were consistently higher (+0.16 to +0.53 Mg ha-1) in strip intercropping than in monoculture when calculated across the entire strips. Soybean yields were consistently lower (-0.02 to -0.14 Mg ha-1) in strips than in monoculture. At the irrigated site, corn yields were also consistently higher (+0.4 to +0.90 Mg ha-1) and soybean yields consistently lower (-0.01 to -0.21 Mg ha-1) in the strip-intercropping system compared with monoculture. At the rainfed site, grain sorghum yields were higher (+0.35 to +0.40 Mg ha-1) and soybean yields inconsistent (-0.08 to +0.01 Mg ha-1) in strip intercropping compared with monoculture. At the irrigated site, grain sorghum yields were higher (+0.28 to +0.62 Mg ha-1) and soybean yields inconsistent (-0.02 to +0.02 Mg ha-1) in strip intercropping compared with monoculture. The significant differences between inside and border row positions are attenuated in the evaluation of whole strips, since there are only two border rows and six inside rows in each strip.
|
In considering the crop yield components and how these could be used to suggest redesign of systems for greater productivity, our results are not consistent. Seed number was more often significantly different between row positions than seed weight, suggesting that the reproductive period is more important in terms of competition for growth resources. In general, there is an increase in seed number in the border rows, especially in corn, and this appears to be due in part to higher plant number in the borders. To further exploit this advantage, the plant density could be increased in the border rows of corn strips. However, the possible increase in reduction of soybean yields may negate the advantage of such a change. Grain sorghum yield components follow the same trend as those of corn, but the differences between border rows and inside rows are less. There is no clear indication from these results how the system should be redesigned, except for the possible use of narrower strips (e.g., alternating six-row or four-row strips), to increase the proportion of border to inside rows and thus exploit the advantage of higher cereal yields in the strip borders.
|
Summary |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionSummaryREFERENCES |
|---|
We conclude that yields and yield components of corn and soybean are more affected by strip intercropping than those of grain sorghum and soybean in similar systems. When strip intercropping and monoculture systems were compared, there was no consistent advantage of the strips as measured by land equivalent ratio, nor were there significant differences in gross economic returns between the two systems. We suggest that strip intercropping could be valuable to reduce soil erosion, if there are appropriate practices and little additional management cost, without changing returns to the cropping systems. The yield component data gave no clear indication of how systems should be redesigned for greater productivity. Measurements of light, nutrient, and water use by individual rows across the strips could reveal more detail on how component crops are affected by these two systems. Different cultivars of the component crops and other cultural practices, such as changing relative plant populations, could be tested to see if they can improve the efficiency of the border rows and thus of the strip-intercropping system. These are areas for future research.
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionSummaryREFERENCES |
|---|
Received for publication June 22, 1998.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionSummaryREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. C. Mason, D. Kathol, K. M. Eskridge, and T. D. Galusha Yield Increase Has Been More Rapid for Maize than for Grain Sorghum Crop Sci., July 1, 2008; 48(4): 1560 - 1568. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Chen, M. Westcott, K. Neill, D. Wichman, and M. Knox Row Configuration and Nitrogen Application for Barley-Pea Intercropping in Montana Agron. J., November 1, 2004; 96(6): 1730 - 1738. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Maman, S. C. Mason, D. J. Lyon, and P. Dhungana Yield Components of Pearl Millet and Grain Sorghum across Environments in the Central Great Plains Crop Sci., November 1, 2004; 44(6): 2138 - 2145. [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 | |||
