Crop Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 19 March 2008
Published in Crop Sci 48:814-822 (2008)
© 2008 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Armstrong, K. L.
Right arrow Articles by Albrecht, K. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Armstrong, K. L.
Right arrow Articles by Albrecht, K. A.
Agricola
Right arrow Articles by Armstrong, K. L.
Right arrow Articles by Albrecht, K. A.
Related Collections
Right arrow Forage Management
Right arrow Maize
Right arrow Other Forage Crops

Effect of Plant Density on Forage Yield and Quality of Intercropped Corn and Lablab Bean

Kevin L. Armstronga,* and Kenneth A. Albrechtb

a Dep. of Crop Sciences, Univ. of Illinois, 1102 S. Goodwin Ave., Urbana, IL 61801
b 1575 Linden Dr., Dep. of Agronomy, Univ. of Wisconsin-Madison, Madison, WI 53706. This research was partially supported by funding through USDA Cooperative State Research Education and Extension Service (CSREES) Hatch Project WIS04802

* Corresponding author (karmstro{at}uiuc.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Low crude protein (CP) concentration in corn (Zea mays L.) silage is its major dietary limitation in dairy rations. In this experiment, four densities of corn (20,000, 40,000, 60,000, and 80,000 plants ha–1) and four densities of lablab bean [Lablab purpureus (L.) Sweet] (0, 40,000, 80,000, 120,000 plants ha–1) were intercropped to determine the optimal planting density in terms of forage yield, nutritive value, estimated milk production, and forage nutrient value. Experiments were conducted near Arlington and Lancaster, WI. Corn was sown in late April and lablab bean was sown in rows 8 cm beside corn rows 2 wk after corn planting. Averaged over locations, forage dry matter (DM) yield ranged from 11 Mg ha–1 to 20 Mg ha–1 as corn plant density increased from 20,000 to 80,000 plants ha–1. Lablab bean increased CP concentration by 22 g kg–1 DM between corn densities of 80,000 plants ha–1 and 20,000 plants ha–1 at a bean density of 120,000 plants ha–1. Calculated milk ha–1 values ranged from 17,500 kg ha–1 to 34,400 kg ha–1 as corn plant density increased from 20,000 to 80,000 plants ha–1. This experiment does not show benefit to addition of lablab bean into high producing corn stands. Corn sown at a density of 80,000 plants ha–1 and 0 bean plants ha–1 is recommended to maximize forage DM yield, milk ha–1, and crop value. Alternatively, addition of lablab bean into low density corn stands did increase CP concentration and feed nutrient value of the forage.

Abbreviations: CP, crude protein • DIP, degradable intake protein • DM, dry matter • EE, ether extract • IVTD, in vitro true digestibility • NDF, neutral detergent fiber • NDFd, neutral detergent fiber digestibility • SECV, standard error of cross validation • UIP, undegradable intake protein


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
CORN SILAGE is a high yielding, palatable forage with high energy density (Coors and Lauer, 2001). Relatively low crude protein (CP) concentration, often in the range of 70 to 80 g kg–1 (Darby and Lauer, 2002), is its major limitation in dairy rations. Intercropping corn with legumes has been explored as a means to increase CP concentration in silage (Armstrong et al., 2007; Bryan and Materu, 1987; Herbert et al., 1984; Kaiser and Lesch, 1977). Herbert et al. (1984) reported an average 29% increase in CP concentration and similar yield of forage when intercropping soybean [Glycine max (L.) Merr.] and corn in a 1:1 ratio compared to monoculture corn.

Many factors influence the performance of mixtures and individual components when two species are grown together. As outlined by Francis (1986), genotype, sowing date, crop density, crop arrangement, and nutrient regimes have potential to influence performance of intercropping systems. Kaiser and Lesch (1977) proposed that the density of corn and a legume crop would heavily influence the forage yield and species composition of corn–bean mixtures. In studies where densities of climbing beans and corn have been altered, CP concentration and forage yield have been affected (Bryan and Materu, 1987; Kaiser and Lesch, 1977).

In an experiment by Kaiser and Lesch (1977) conducted in South Africa, varying densities of corn and lablab bean were used to determine effects on dry matter (DM) yield, CP, and legume proportion in the mixtures. At a constant lablab bean density of 108,000 plants ha–1 and corn plant densities ranging from 72,000 to 18,000 plants ha–1, CP concentration increased by 44% and total DM yield was decreased by 28% at lower densities. Similarly, Bryan and Materu (1987) in West Virginia found that an intercrop of corn (64,600 plants ha–1) and P. vulgaris (215,200 plants ha–1) reduced corn DM yield by 26% and increased mixture CP concentration by 16% compared to monoculture corn at the same plant density. However, the intercrop system of corn and cowpea [Vigna unguiculata (L.) Walp] did not significantly reduce corn DM yield and increased total DM yield by 7% and CP concentration by 9%. Bryan and Materu (1987) concluded that cowpea was the best intercrop with corn because of increased mixture CP concentration and similar total DM yield to monoculture corn.

Lablab bean accounted for highest bean concentration in mixture (114 g kg–1) compared to other climbing beans, and increased CP concentration by 13% compared to monoculture corn across four WI environments (Armstrong et al., 2007). It is not known, however, how different densities of lablab bean and corn interact for mixture yield and nutritive value in a high-yield upper Midwest environment. This experiment was designed to determine optimal plant densities of intercropped corn and lablab bean for forage yield, nutritive value, estimated milk production, and forage nutrient value.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Field experiments were conducted in 2005 at the University of Wisconsin Agricultural Research Stations near Arlington (43°18' N, 89°21' W) and Lancaster (42°50' N, 90°47' W), WI. The experiment at Arlington was conducted on Plano silt loam (fine-silty, mixed, mesic Typic Argiudoll), on a relatively flat and well-drained field and at Lancaster on Rozetta silt loam (fine-silty, mixed, superactive, mesic Typic Hapludalfs), on a slightly sloping well-drained field. The previous crop at both sites was soybean. Before tillage, 135 kg ha–1 N and 180 kg ha–1 N as urea was applied at Arlington and Lancaster, respectively. Tillage operations included chisel plowing and field cultivation. Soil fertility levels at both locations were maintained at optimal levels for corn silage production (Kelling et al., 1998). Corn hybrid DKC 50–20 was planted on 25 April at Arlington and 26 April at Lancaster at 93,000 seeds ha–1 using a commercial corn planter and later hand-thinned to designated corn density treatments. Permethrin [(3-phenoxyphenyl) methyl( ± )cis-trans 3-(2,2-dichloroethenyl)-2,2-dimethylcyclopropanecarboxylate] (6 g ha–1 a.i.), for control of corn wireworm [Melanotus communis (Gyll.)] and seed corn maggot [Delia platura (Meigen)], and carboxin (5,6-dihydro-2-methyl-N-phenyl-1,4-oxathin-3-carboxamide) (8 g ha–1 a.i.), for control of seed rots and decay, were applied at planting at Lancaster. Flumetsulam [N-(2,6-difluorophenyl)-5-methyl(1,2,4)triazolo(1,5-a)pyrimidine-2-sulfonamide] (28 g ha–1 a.i.) and s-metolachlor + safener [(1S)-2-chloro-N-(2-ethyl-6-methylphenyl)-N-(2-methoxy-1-methylethyl)acetamide] (1.42 kg ha–1 a.i.) were applied pre-emergence to control weeds at both locations.

Experimental treatments of corn density (20,000, 40,000, 60,000, and 80,000 plants ha–1) and ‘Rongai’ lablab bean density (0, 40,000, 80,000, 120,000 plants ha–1) were selected by review of previous research (Armstrong et al., 2007, Bryan and Materu, 1987; Kaiser and Lesch, 1977). Bean seeds were inoculated with appropriate rhizobia (Nitragin Inc., Milwaukee, WI) and sown at 177,900 seeds ha–1 with a four-row John Deere Flex 71 planter (Deere and Company, Moline, IL) about 8 cm to one side of the corn rows and 2 wk after corn planting. When corn plants reached the V6 stage (Ritchie et al., 1992), both corn and bean were hand-thinned to their respective treatment levels.

The experiment was set up as a randomized complete block design with four replications at each location. Each replication consisted of a factorial arrangement of four lablab bean densities and four corn densities and experimental units consisted of four corn rows with associated lablab bean rows in 8.23-m long plots. All four rows in each experimental unit received treatments. One of the middle corn rows with associated lablab bean was harvested for forage.

Near infrared reflectance spectroscopy (NIRS) was used to estimate bean and corn proportions in mixtures. On day of harvest, representative corn and bean plants were removed from one of the middle rows with treatments of 80,000 plants ha–1 corn/40,000 plants ha–1 bean, 20,000 plants ha–1 corn/120,000 plants ha–1 bean, 40,000 plants ha–1 corn/40,000 plants ha–1 bean and separated into corn and bean components. Separated corn and bean plants were dried at 60°C and ground with a Christy hammer mill (Christy, Suffolk, UK) equipped with a 1-mm screen. Pure fractions (48) and mixtures (36) of corn and bean created by combining the pure fractions were used for NIRS equation development. Samples were scanned with a NIRSystems 6500 near infrared reflectance spectrophotometer (FOSS NIRSystems Inc., Eden Prairie, MN) equipped with a ring cup autosampler. Near-infrared reflectance spectra (1/R) were obtained between the wavelengths of 400 and 2498 nm. Data management and equation development were performed using WinISI 1.50 (Infrasoft Int. Limited, Port Matilida, PA). Calibration statistics [coefficient of determination (R2) and standard error of cross validation (SECV)] for determining bean concentration in validation mixtures were SECV = 18 (R2 = 0.99) for one equation developed over both locations (Martens and Naes, 1989; Shenk and Westerhaus, 1991, 1994). The equation was used to predict bean concentration in mixtures and corn DM mass was determined by subtracting bean DM mass from total plot DM mass.

One middle row of each plot was harvested at the 50% kernel milk line stage on 7 September at Lancaster and 15 September at Arlington. Harvest rows were chopped to a theoretical cutting length of 1 cm with a small, commercial forage harvester and a 1-kg subsample was dried at 60°C to determine DM concentration of the forage. The dried subsample was ground with a hammer mill equipped with a 1-mm screen.

Forages were analyzed for total N concentration by the Dumas method (AOAC, 1990) with an automated analyzer (LECO Model FP-528; LECO Corp., St. Joseph, MI). Crude protein concentrations were calculated by multiplying total N by 6.25. Neutral detergent fiber (NDF) concentrations were determined by the batch procedure outlined by ANKOM Technology Corp. (Fairport, NY). Subsamples (0.25 g each) were analyzed for in vitro true digestibility (IVTD) using rumen fluid from a lactating Holstein cow on a total mixed ration and buffer solution described by Goering and Van Soest (1970) with the Daisy II200 in vitro incubator and the ANKOM200 fiber analyzer (ANKOM Technology Corp., Fairport, NY). Neutral detergent fiber digestibility (NDFd) was calculated from the NDF and IVTD values as 100{[NDF-(100-IVTD)]/NDF} (data not shown but used in milk production models). Ash concentration was determined by combustion of a 1.0-g subsample at 600°C for 2 h (data not shown but used in milk production models). Starch concentrations were determined by the procedures of Rong et al. (1996) and Owens et al. (1999).

Potential milk production estimates were calculated with the MILK2000 spreadsheet (Schwab et al., 2003). Milk Mg–1 forage DM and milk ha–1 were calculated for corn and mixture forages. Values for ether extract (EE) were estimated from weighted values of corn silage (National Research Council, 2001) and lablab bean (Díaz et al., 2003) (depending on bean percentage in mixtures) while neutral detergent insoluble CP values were estimated from weighted values of corn silage and alfalfa in the NRC tables (National Research Council, 2001).

Forage nutrient values were calculated with FEEDVAL4, a spreadsheet developed to assign a dollar value to feed ingredients (Howard and Shaver, 1997). The term forage nutrient value refers to the output of the spreadsheet, which allows the user to compare feeds based on current prices of feed ingredients and determine cost effectiveness. The FEEDVAL4 spreadsheet uses blood meal (undegradable intake protein, UIP), urea (degradable intake protein, DIP), shelled corn (energy), tallow (fat), dicalcium phosphate (phosphorus), and calcium carbonate (calcium) as reference feed ingredients. Prices for reference ingredients were based on April 2006 market values. The DM and CP components were measured values while the total digestible nutrients (TDN) of the mixtures was calculated using MILK2000. The UIP and DIP percent of CP were estimated from a combination of corn silage and alfalfa (Medicago sativa L.) values according to Linn et al. (1994), while the fat concentration was estimated from corn silage (National Research Council, 2001). Calcium and phosphorus concentrations for corn and lablab bean were estimated from corn silage and alfalfa values from the NRC tables (National Research Council, 2001). Feed nutrient values were determined for 1 Mg of DM of each mixture and then multiplied by the corresponding mixture yield to provide crop value for 1 ha of each mixture.

Data from both locations were pooled and analyzed with the Windows version of SAS software package release 9.1. Tests concerning heterogeneity of variances were conducted to assess the appropriateness of pooling the data; however, no such problems existed in subsequent models. The STEPWISE selection option in the REG procedure (SAS Institute, 2002) was used to select significant (entry probability level of 0.10, exit probability level of 0.15) model coefficients with the response variables of mixture yield, forage composition, calculated milk Mg–1 and ha–1, and forage nutrient values. Linear, quadratic, and all possible interaction effects of both corn and bean density were considered as model terms. Regression models were evaluated using adjusted R2 values, and Mallows' Cp (Mallows, 1973). Significant coefficients were then graphed using the G3D or GPLOT procedures (SAS Institute, 2002) depending on whether coefficients exhibited response surface or 2-dimensional structure. The MIXED procedure (SAS Institute, 2002) was used to detect treatment differences for nutritive values of pure corn and bean subsamples (data not shown). Treatments were considered fixed while blocks nested within locations were considered random effects. Single degree of freedom contrasts were used to explore differences (P < 0.05) between treatments. The Pearson correlation coefficient, calculated in the CORR procedure (SAS Institute, 2002), was used to detect correlations between bean concentration in mixtures and mixture CP, NDF, IVTD, and starch concentrations.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Environment
Temperatures were lower than normal in May, but higher than normal in June, at both Arlington and Lancaster sites (Table 1 ). Arlington received less than normal precipitation in June and early July (Table 1), causing corn to show signs of stress, so the site was irrigated with 51 mm water on 28 June, 11 July, and 19 July to supplement rainfall to near normal levels. These conditions were suitable for excellent early development of both corn and lablab bean. In these two Wisconsin environments, day length was too long for flower development and subsequent seed production in lablab bean. However, in environments where seed production is possible, nutritive value results will differ. Hintz et al. (1992) concluded that seed makes a substantial contribution to the feeding value of soybean harvested for forage.


View this table:
[in this window]
[in a new window]

  		Table 1. Mean monthly precipitation and temperature at the Arlington and Lancaster, WI Research Stations in the 2005 growing season.

 
Forage Yield and Nutritive Value
Forage mixture DM yields were primarily influenced by the linear effect of corn plant density (98% of total adjusted R2 value) (Table 2 ). Forage yield ranged from 11 Mg ha–1 to 20 Mg ha–1 as corn plant density increased from 20,000 to 80,000 plants ha–1 (Fig. 1 ). Bean density did not have a detectable effect on total mixture yield as bean density increased from 0 to 120,000 plants ha–1, even at the low corn density (data not shown). These results indicate that forage yield is largely determined by the yield of the corn plants and that lablab bean plants do not compensate for reduced corn yield at low density. Bryan and Materu (1987) found similar results in which corn–P. vulgaris mixture DM yields were lower with a corn density of 40,400 plants ha–1 compared to a corn density of 50,500 plants ha–1. Kaiser and Lesch (1977) also found that as corn densities increased from 18,000 to 72,000 plants ha–1, with a constant lablab bean density of 108,000 plants ha–1, DM yield for the corn–lablab bean intercrop increased by 39%. But maximum yield in both of these environments was about half of that observed in the current research, suggesting some environmental limitations to corn production.


View this table:
[in this window]
[in a new window]

  		Table 2. Polynomial model estimates for forage yield and bean concentration of corn-lablab bean mixtures grown in two environments.{dagger}

 

Figure 1
View larger version (7K):
[in this window]
[in a new window]

  		Figure 1. Response of total forage yield to density of corn. Data are pooled over four lablab bean densities and two environments.

 
Corn DM yield was influenced primarily by the linear effects of corn (91% of total adjusted R2 value) and bean (6% of total adjusted R2 value) plant densities (Table 2). Corn DM yields ranged from 7 to 20 Mg ha–1 as corn plant density increased from 20,000 to 80,000 plants ha–1, depending on bean density (Fig. 2 ). As bean density increased, corn yield decreased. For example, at a corn density of 80,000 plants ha–1, corn yield decreased by 2.5 Mg ha–1 as bean density increased from 0 to 120,000 plants ha–1 (Fig. 2). The response curve is parabolic upward as bean density increases, which suggests that corn yields were reduced most near 85,000 bean plants ha–1 and that increasing the bean density to 120,000 plants ha–1 did not further negatively affect corn yield (Fig. 2). These data suggest that increasing the bean density up to 85,000 plants ha–1 reduces corn yield in mixture. Because lablab bean is a climbing, vining plant, its broad leaves are placed at or near the top of the mixture canopy through most of the growing season making it an efficient competitor for light. Maasdorp and Titterton (1997) found that in a corn–velvet bean mixture that contained 30% bean, corn production was 4.1 Mg ha–1, compared to 8.0 Mg ha–1 for monoculture corn. Bryan and Materu (1987) also reported reduction in corn yield from 8.8 Mg ha–1 in monoculture corn to 6.5 Mg ha–1 when intercropped with climbing P. vulgaris with a corn density of 64,600 plants ha–1 and a bean density of 215,200 plants ha–1.


Figure 2
View larger version (58K):
[in this window]
[in a new window]

  		Figure 2. Response of corn yield to densities of corn and lablab bean in mixtures. Data are pooled over two environments.

 
Bean concentration in mixtures ranged from 0 to 385 g kg–1 DM and was influenced mainly by the linear effect of bean density (42% of total adjusted R2 value) and the interaction of the linear effect of corn density and quadratic effect of bean density (42% of total adjusted R2 value) (Table 2). Bean concentration increased with bean density across all corn densities, but the response curve shows a pronounced difference in how bean concentration responds to low and high corn densities (Fig. 3 ). For example, at a corn density of 80,000 plants ha–1 and a bean density of 120,000 plants ha–1, the bean concentration was 93 g kg–1 DM. In comparison, at a corn density of 20,000 plants ha–1 and a bean density of 120,000 plants ha–1, the bean concentration was 340 g kg–1 DM. The shape of the response surface shows that bean concentration is maximized at a bean density of 89,000 plants ha–1, and that increasing bean density to 120,000 plants ha–1 does not increase bean concentration in the harvested mixture (Fig. 3). These data suggest that regardless of corn density, bean concentration can be increased in the harvested mixtures by increasing bean density. Bean concentration is higher in the low corn densities because of less competition for nutrients and light and greater bean biomass production. Bryan and Materu (1987) also found that as P. vulgaris densities increased from 20,200 to 80,800 plants ha–1, mixture bean concentration more than doubled. Kaiser and Lesch (1977) saw that a constant lablab bean density of 108,000 plants ha–1 and corn plant densities ranging from 18,000 to 72,000 plants ha–1 decreased bean concentration from 408 to 117 g kg–1 DM.


Figure 3
View larger version (31K):
[in this window]
[in a new window]

  		Figure 3. Response of bean concentration to densities of corn and lablab bean in mixtures. Data are pooled over two environments.

 
The CP concentration in mixtures was mainly affected by the linear effects of corn (89% of total adjusted R2 value) and bean (6% of total adjusted R2 value) densities (Table 3 ). Crude protein concentration ranged from 53 to 83 g kg–1 DM as corn density decreased from 80,000 plants ha–1 to 20,000 plants ha–1 and bean density increased from 0 to 120,000 plants ha–1 (Fig. 4 ). The response surface acts differently between high and low corn densities. For example, at a constant bean density of 120,000 plants ha–1 and corn densities ranging from 80,000 plants ha–1 to 20,000 plants ha–1, the CP concentration increased by 22 g kg–1 DM. The curve shows a maximum CP concentration at a corn density of 20,000 plants ha–1 and a bean density of 80,000 plants ha–1. The CP response curve follows a similar pattern to the bean concentration curve (Fig. 3) in that as bean density is increased, bean concentration and CP increase (Pearson correlation, r = 0.62; P < 0.0001; data not shown). Pure lablab bean had a higher CP concentration (106 g kg–1 DM) than monoculture corn (65 g kg–1 DM) (data not shown); thus, increasing the proportion of legume in the mixture would increase CP concentration. The CP curve is lower at high corn densities because bean proportions were lower and did not contribute to increasing the total mixture CP concentration. Kaiser and Lesch (1977) found that mixture CP concentration increased by 44% at a constant lablab bean density of 108,000 plants ha–1 and corn plant densities ranging from 72,000 to 18,000 plants ha–1. Bryan and Materu (1987) found that intercropping cowpeas and corn increased mixture CP concentration by 9% and did not lower total DM yield compared to monoculture corn.


View this table:
[in this window]
[in a new window]

  		Table 3. Polynomial model estimates for nutritive value of corn-lablab bean mixtures grown in two environments.{dagger}

 

Figure 4
View larger version (18K):
[in this window]
[in a new window]

  		Figure 4. Response of CP concentration to densities of corn and lablab bean in mixtures. Data are pooled over two environments.

 
The NDF concentration was affected by the linear effects of bean (51% of total adjusted R2 value) and corn (27% of total adjusted R2 value) densities, along with the quadratic effects (11% of total adjusted R2 value each) of corn and bean (Table 3). Neutral detergent fiber concentrations ranged from 349 to 428 g kg–1 DM as bean density increased from 0 to 120,000 plants ha–1 and corn density decreased from 80,000 to 20,000 plants ha–1 (Fig. 5 ). The response curve differs for high and low densities of corn. At a constant bean density of 120,000 plants ha–1 and corn densities ranging from 20,000 to 80,000 plants ha–1, the NDF concentration decreased by 31 g kg–1 DM (Fig. 5). The NDF response curve (Fig. 5) along with the bean concentration (Fig. 3) and CP response curves (Fig. 4) show that increased bean density leads to increased bean, CP, and NDF concentrations in the mixtures. Bean concentration in mixtures had a significant (P < 0.0001) Pearson correlation coefficient of r = 0.62 to NDF concentration in mixtures (data not shown). Like the CP response curve (Fig. 4), increase in NDF concentration is not as great at high corn densities because the proportion of bean in mixtures was lower. At lower corn densities the bean proportion was greater and the NDF concentration increased more. The NDF response curve is maximized near a bean density of 98,000 plants ha–1, after which greater bean density does not increase NDF concentration (Fig. 5). The increase in mixture NDF concentration by increasing bean density is at least partially driven by pure lablab bean having a higher NDF concentration (403 g kg–1 DM) than monoculture corn (382 g kg–1 DM) (data not shown).


Figure 5
View larger version (28K):
[in this window]
[in a new window]

  		Figure 5. Response of NDF concentration to densities of corn and lablab bean in mixtures. Data are pooled over two environments.

 
The IVTD concentration is mainly affected by the linear effect of bean density (60% of total adjusted R2 value) and the interaction of the linear effect of corn density and quadratic effect of bean density (28% of total adjusted R2 value) (Table 3). The IVTD concentrations ranged from 782 to 832 g kg–1 DM as bean density decreased from 120,000 to 0 plants ha–1 and corn density increased from 20,000 to 80,000 plants ha–1 (Fig. 6 ). The IVTD response curve shows the opposite response to plant density compared to the bean (Fig. 3), CP (Fig. 4), and NDF (Fig. 5) concentration curves, in that as bean density increased, IVTD concentration declined. Bean concentration in mixtures had a significant (P < 0.0001) Pearson correlation coefficient of r = –0.63 to IVTD concentration in mixtures (data not shown). The IVTD response surface has a parabolic curve upward as bean density increases, with different localized minimum IVTD concentrations for the middle corn densities tested. However, the absolute minimum IVTD concentration (782 g kg–1 DM) occurs at a bean density of 120,000 plants ha–1 and a corn density of 20,000 plants ha–1 (Fig. 6). The decrease in IVTD concentration as bean density increases is at least partially due to pure lablab bean having a lower IVTD concentration (780 g kg–1 DM) compared to monoculture corn (831 g kg–1 DM) (data not shown).


Figure 6
View larger version (23K):
[in this window]
[in a new window]

  		Figure 6. Response of IVTD concentration to densities of corn and lablab bean in mixtures. Data are pooled over two environments.

 
Starch concentration was affected largely by the linear effects of corn (56% of total adjusted R2 value) and bean (25% of total adjusted R2 value) densities, along with quadratic effects (8 and 7% of total adjusted R2 value, respectively) of corn and bean (Table 3). Starch concentration ranged from 412 to 256 g kg–1 DM as bean density increased from 0 to 120,000 plants ha–1 and corn density declined (Fig. 7 ). Starch concentration response differed between high and low corn densities (Fig. 7). At a constant bean density of 120,000 plants ha–1 and corn densities ranging from 20,000 to 80,000 plants ha–1, the starch concentration increased by 98 g kg–1 DM (Fig. 7). Starch concentrations diminish as the curve approaches a corn density of 20,000 plants ha–1, with the lowest starch concentration (256 g kg–1 DM) occurring at a bean density of 92,000 plants ha–1. Bean concentration in mixtures had a significant (P < 0.0001) Pearson correlation coefficient of r = –0.78 to starch concentration in mixtures (data not shown). In corn harvested for silage, most of the starch is found in the grain, but some is also found in the stover (Albrecht et al., 1986). Corn kernel development requires remobilization and translocation of starch in the stover to fill the sink created by corn kernels (Coors et al., 1997). According to Coors et al. (1994), grain can account for 50% of corn silage DM yield during a normal year. Starch concentration in vegetative legume forage is much lower than in corn, often ranging from 2 to 48 g kg–1 DM in fresh alfalfa (Owens et al., 1999). Our results indicate that a higher starch concentration occurs with a higher corn density and lower bean density. Lower starch concentration in the corn–bean mixtures may be due to a lower proportion of grain, especially at low corn densities, because of competition with the beans for light and other resources. The difference in starch concentration of legume forage compared to corn may also contribute to a low starch concentration in the low-density corn and high-density bean mixtures.


Figure 7
View larger version (31K):
[in this window]
[in a new window]

  		Figure 7. Response of starch concentration to densities of corn and lablab bean in mixtures. Data are pooled over two environments.

 
Milk Mg–1 forage integrates laboratory measures of forage quality and is calculated from the MILK2000 spreadsheet (Schwab et al., 2003). Milk Mg–1 forage was mainly affected by the linear effect of bean density (36% of total adjusted R2 value) and the interaction of the linear effect of corn density and the quadratic effect of bean density (45% of total adjusted R2 value) (Table 4 ). Milk Mg–1 forage ranged from 1790 to 1500 kg Mg–1 as bean density increased from 0 to 120,000 plants ha–1 and corn density declined from 80,000 to 20,000 plants ha–1 (Fig. 8 ). The response of milk Mg–1 forage differed among corn density treatments (Fig. 8). At a constant bean density of 120,000 plants ha–1 and corn densities ranging from 20,000 to 80,000 plants ha–1, the calculated milk Mg–1 forage increased by 139 kg Mg–1. The response surface is a parabolic curve upward as bean density increases, with different localized minimum milk Mg–1 forage values for the middle corn densities tested. However, the absolute minimum milk Mg–1 forage (1500 kg Mg–1) occurs at a bean density of 120,000 plants ha–1 and a corn density of 20,000 plants ha–1 (Fig. 8). These data further show that the highest calculated milk Mg–1 forage occurs at 0 plants ha–1 bean density, regardless of corn density. Cox and Cherney (2005) reported 1692 kg milk Mg–1 forage, averaged over three corn hybrids in New York. However, the corn hybrids reported by Cox and Cherney (2005) had higher NDF and lower starch concentrations compared to the current research, which resulted in lower calculated milk production. The response curve does not continue to decline with corn densities of 40,000 to 60,000 plants ha–1 and a high bean concentration of 120,000 plants ha–1 (Fig. 8). This may be explained because the increased bean concentration provided a higher CP concentration, which ultimately contributed as additional predicted energy and milk Mg–1 forage. Calculated milk Mg–1 is greater at higher corn densities because of high grain content and digestibility of corn which contributes to greater energy density of forage.


View this table:
[in this window]
[in a new window]

  		Table 4. Polynomial model estimates for milk estimates and forage nutrient values of corn-lablab bean mixtures grown in two environments.{dagger}

 

Figure 8
View larger version (29K):
[in this window]
[in a new window]

  		Figure 8. Response of calculated milk Mg–1 to densities of corn and lablab bean in mixtures. Data are pooled over two environments.

 
Calculated milk ha–1 integrates forage quality and yield calculated from the MILK2000 spreadsheet (Schwab et al., 2003). Milk ha–1 is mainly influenced by the linear effect of corn density (97% of total adjusted R2 value) (Table 4). Calculated milk ha–1 values ranged from 17,900 kg ha–1 to 34,500 kg ha–1 as corn plant density increased from 20,000 to 80,000 plants ha–1 (Fig. 9 ). Bean density did not have a detectable effect on calculated milk ha–1 as bean density increased from 0 to 120,000 plants ha–1, even at the lowest corn density. These results indicate that calculated milk ha–1 is largely determined by the yield of the corn plants and that lablab bean plants do not compensate for reduced milk ha–1 at low corn density. For comparison, Cox and Cherney (2005) reported 25,700 kg calculated milk ha–1 averaged over three hybrids. However, average DM yield for the same hybrids was 14.9 Mg ha–1 (Cox and Cherney, 2005), approximately 25% lower than in the current research and the primary reason for lower calculated milk yields. Because calculated milk ha–1 is a variable driven by forage yield and quality, any reduction in total DM yield reduces the predicted milk ha–1 values. Milk ha–1 follows a very similar pattern to forage yield (Fig. 1).


Figure 9
View larger version (8K):
[in this window]
[in a new window]

  		Figure 9. Response of calculated milk ha–1 to density of corn. Data are pooled over four lablab bean densities and two environments.

 
The calculations and equations for the MILK2000 spreadsheet are driven by forage fiber characteristics, energy availability of all chemical constituents, and yield. The model does not take into account the added value of additional CP in the ration so FEEDVAL4 (Howard and Shaver, 1997) was used to evaluate this characteristic of the corn–bean mixtures. Feed nutrient value and crop value (each components of forage nutrient value) were mainly influenced by the linear effect of corn density (100% and 97% of total adjusted R2 value respectively) (Table 4). The feed nutrient value in $ Mg–1 DM declined linearly as corn density increased from 20,000 to 80,000 plants ha–1 (Fig. 10 ). The crop value ($ ha–1), based on value of feed ingredients, increased as corn density increased (Fig. 11 ). The increased feed nutrient value in $ Mg–1 DM is attributed to increased CP concentration in the low corn density (20,000 plants ha–1) (Fig. 4) and increasing bean proportions (Fig. 3). FEEDVAL4 estimates the value of feeds due to the price of different nutrients with CP being a major contributor. However, since the crop value in $ ha–1 takes into account total mixture yield, the value of the feed becomes a combination of $ Mg–1 DM and yield. The highest crop value ($1500 ha–1) came from a corn density of 80,000 plants ha–1 in our experiment, without regard to bean density, and is associated with the greatest corn forage yield (Fig. 11).


Figure 10
View larger version (7K):
[in this window]
[in a new window]

  		Figure 10. Response of calculated feed nutrient value to density of corn. Data are pooled over four lablab bean densities and two environments.

 

Figure 11
View larger version (7K):
[in this window]
[in a new window]

  		Figure 11. Response of calculated crop value to density of corn. Data are pooled over four lablab bean densities and two environments.

 

   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Results from this experiment show that forage yield of monoculture corn and corn–lablab bean mixtures is largely determined by corn density and that addition of beans does not increase DM yield of mixtures compared to monoculture corn at any density tested. Bean concentration in mixtures, along with CP concentration, was maximized at low corn and high bean densities. The NDF concentration increased while the IVTD concentration decreased as bean concentration increased in mixtures. Furthermore, the starch concentration of mixtures declined from a corn density range of 60,000 to 20,000 plants ha–1 and a bean density range of 80,000 to 120,000 plants ha–1. The response of calculated milk ha–1 was similar to total mixture DM yield in that it declined as corn density declined, regardless of bean density. The highest crop value ($1500 ha–1) came from a corn density of 80,000 plants ha–1 regardless of bean density. This experiment shows no advantage to addition of bean into high producing corn stands. Corn sown at a density of 80,000 plants ha–1 and 0 bean plants ha–1 is recommended to maximize forage DM yield, milk ha–1, and crop value. Alternatively, addition of bean into low density corn stands increased CP concentration and feed nutrient value ($ Mg–1 DM).


   ACKNOWLEDGMENTS
 
The authors thank Ed Bures for technical assistance in the field and laboratory and Joe Lauer, Pat Flannery, and Tim Wood for assistance in the field.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Armstrong, K. L.
Right arrow Articles by Albrecht, K. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Armstrong, K. L.
Right arrow Articles by Albrecht, K. A.
Agricola
Right arrow Articles by Armstrong, K. L.
Right arrow Articles by Albrecht, K. A.
Related Collections
Right arrow Forage Management
Right arrow Maize
Right arrow Other Forage Crops

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal 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