Agronomy Journal Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 13 July 2005
Published in Agron J 97:1164-1171 (2005)
DOI: 10.2134/agronj2003.0279
© 2005 American Society of Agronomy
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 Similar articles in ISI Web of Science
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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chikoye, D.
Right arrow Articles by Ogunyemi, S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Chikoye, D.
Right arrow Articles by Ogunyemi, S.
Agricola
Right arrow Articles by Chikoye, D.
Right arrow Articles by Ogunyemi, S.
Related Collections
Right arrow Weed Management
Right arrow Agricultural Pesticides
Right arrow Cover Crops

Integrated Weed Management

Integrated Management of Cogongrass [Imperata cylindrica (L.) Rauesch.] in Corn Using Tillage, Glyphosate, Row Spacing, Cultivar, and Cover Cropping

David Chikoyea,*, Udensi E. Udensia and Shola Ogunyemib

a International Institute of Tropical Agriculture (IITA), Oyo Road, Ibadan, Nigeria
b Dep. of Crop Protection, Univ. of Ibadan, Ibadan, Nigeria

* Corresponding author (d.chikoye{at}cgiar.org)

Received for publication November 11, 2003.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Cogongrass [Imperata cylindrica (L.) Raeusch.] is a major weed in the tropics where land is intensively cultivated and fallow duration is less than 5 yr. Field studies were conducted in 1999 and 2000 in Nigeria to evaluate cogongrass response to combinations of five weed control options. Treatment combinations were tillage (hoe tillage and no-tillage), corn (Zea mays L.), row spacing (50 and 75 cm), corn cultivar [open-pollinated (OP) and hybrid], herbicide (glyphosate and no glyphosate), and cover crop {velvetbean [Mucuna cochinchinensis (Lour) A. Chev] and no velvetbean}. The treatment combinations were arranged in a randomized complete block design with three replications. Integrating tillage, herbicide, and cover cropping with velvetbean gave optimal control of cogongrass. Corn height and leaf area were negatively correlated with cogongrass shoot, rhizome, and total biomass. Corn grain yield was negatively correlated with cogongrass shoot biomass and total cogongrass biomass (shoot + rhizome). Good control of other weeds was achieved through the use of narrow corn row spacing and cover cropping with velvetbean. Tillage, narrow corn row spacing, and the use of herbicide had a positive effect on corn grain yield. The use of competitive cultivars, narrow row spacing, cover crop, and herbicide may be sustainable approaches to the control of cogongrass and other weeds in corn.

Abbreviations: LAI, leaf area index • OP, open-pollinated • PAR, photosynthetically active radiation • TPAR, photosynthetically active radiation transmitted through the canopy • WAP, weeks after planting


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
COGONGRASS is a rhizomatous, perennial grass weed, widely distributed throughout the tropics. It may also be found in some warm areas of the temperate region (Holm et al., 1977). It is a serious problem in southeastern Asia; some parts of the United States, notably in Florida; and in the moist savanna of West Africa (Wilcut et al., 1988; Garrity et al., 1997). The deleterious effects of cogongrass include infesting land, competing with crops, injuring humans and animals, and being a fire hazard in perennial crops, especially when fire is used to hunt for small ruminants for food (Townson, 1991; Terry et al., 1997; Chikoye et al., 1999). Crop production in the savanna zone of West Africa has not reached its maximum potential, partly due to the problems associated with cogongrass infestation. The difficulty of control and the use of inefficient indigenous options, as well as a shortage of labor and the high cost of weeding, force farmers to abandon fertile arable land in favor of less infested marginal land (Akobundu et al., 2000).

Uncontrolled cogongrass can cause tremendous losses in major crops of West Africa. For example, yield reduction attributed to cogongrass interference can be as high as 70% in corn (Udensi et al., 1999), 78% in yam (Dioscorea spp.) and cassava (Manihot esculenta Crantz) (Koch et al., 1990), and 40% in soybean [Glycine max (L.) Merr.] (Avav, 2000). Complete crop failure usually occurs when crops are grown in plots slashed during land preparation, without additional weeding during the growth cycle (Udensi et al., 1999). Slashing, usually performed using machetes, only suppresses the shoots with little or no effect on the rhizomes of cogongrass (Townson, 1991; Chikoye et al., 1999; Udensi et al., 1999).

A period of undisturbed fallow of up to 5 yr can reduce the population or biomass of cogongrass by 70% and may guarantee farmers 2 to 5 yr of cropping before cogongrass reinfestation (Okigbo, 1982). Long fallow periods are no longer possible, owing to the increasing human population pressure on limited arable land (Siebert and Kuncoro, 1987; Chikoye et al., 1999). Other interventions include hoe weeding, tillage, slashing, burning, chemical control, alley cropping, and the use of cover crops. Hoe weeding and slashing are the methods most commonly used by small-scale farmers to control cogongrass; weeding is effective only if repeated at least five times per season in corn- or cassava-based cropping systems (Udensi et al., 1999; Chikoye et al., 2001). Slashing, followed by burning, is very effective for removing cogongrass foliage but also has a limited effect on the rhizomes, which rapidly regenerate new shoots (Akobundu et al., 2000; Avav, 2000; Chikoye et al., 2001). Hoe weeding is labor intensive and can consume at least 70% of the total labor budget (Chikoye et al., 2002). Tillage to a depth of 30 cm, or more, is effective against rhizomes (Ivens, 1980). Most small-scale farmers cannot afford this option because of lack of money or shortage of labor. Chemical control is cheaper, faster, and more effective than hoe weeding or slashing (Chikoye et al., 2002). For example, glyphosate [N-(phosphonomethyl)glycine] and imazapyr {2-[4,5-dihydro-4-methyl-4-(1-methylethyl)-5-oxo-1H-imidazol-2-yl]-3-pyridinecarboxylic acid} can suppress cogongrass for 3 to 12 mo, depending on the rate of application and timing (Ivens, 1980; Onyia, 1997). In West Africa, there is limited availability of these herbicides because of poor markets. Farmers perceive chemical control to be very expensive, as one may require making multiple applications (Ivens, 1975; Akobundu et al., 2000). Chemical control may have limited potential in multiple cropping systems involving cereals and broad-leaved crops because not all the crop components of the mixture may be tolerant to a given herbicide.

Alley cropping, which simulates the traditional shifting cultivation, has been reported to control cogongrass effectively. Anoka et al. (1991) reported that shading from lead tree [Leucaena leucocephala (Lam.) de Wit] and quick stick (Gliricidia sepium Jacq.) reduced cogongrass rhizome biomass by >90% over 2 yr. Despite the numerous advantages of alley cropping, the technology has not yet gained wide acceptance by farmers. Factors that prevent adoption include the additional labor needed for establishment, pruning, and general management of the alley species rows. Many researchers have shown the beneficial effect of cover crops, e.g., velvetbean, for cogongrass control, soil improvement, and the production of livestock fodder (Hulugalle et al., 1986; Smith et al., 1987; Sanginga et al., 1996; Versteeg et al., 1998; Vissoh et al., 1998; Udensi et al., 1999; Akobundu et al., 2000; Chikoye and Ekeleme, 2001). However, cover crops and the control options discussed above do not provide lasting control. Some of the problems associated with the use of cover crops are the variable effects on cogongrass due to poor growth. Poor performance of some cover crops, e.g., velvetbean, in some areas of West Africa is due to a slow rate of germination, pest attack, and sensitivity to water logging (Carsky et al., 1998; Akobundu et al., 2000; Chikoye and Ekeleme, 2001; Chikoye et al., 2001).

Sustainable management of cogongrass requires an integrated approach that considers crop competitiveness, the biology of the weed, and control options. Swanton and Weise (1991) suggested that better use of crop density and narrow rows would make crops more competitive against weeds. Studies have shown that, when row spacing is reduced or seeding rate increased, canopy development is hastened, and this subsequently increases the crop's ability to compete for incoming photosynthetic radiation (Cox et al., 2001; Maddonni et al., 2001). Cogongrass has a C4 photosynthetic pathway (C4 plants are adapted to grow in an open environment) and, therefore, requires high levels of irradiance and is sensitive to shading (Patterson, 1980). Little is known about the effect of crop cultivar competitiveness on cogongrass; most of the available cogongrass control options have been tested individually and not in an integrated manner. The objective of this study was to investigate the influence of tillage, corn row spacing, corn cultivars, herbicide, and cover cropping on cogongrass in corn.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Field experiments were conducted on farm in 1999 and 2000 on a site infested with cogongrass in Ibadan, Nigeria (7°35' N, 3°55' E) in the forest/savanna transition zone. This zone has average annual temperature of 26°C and receives annual rainfall of 1200 to 1500 mm from April through October. The rainfall has a bimodal distribution pattern, with major peaks occurring in July and September. The soil type was a loamy sand (Oxic Paleustalf) with a pH of 5.9, organic matter ≤ 2%, and soil texture 90% sand, 6% clay, and 4% silt. Before the study, the site was in fallow for 10 yr, occasioned by annual bush burning, because of poor crop yields due to cogongrass infestation.

The experiment was a randomized complete block design with three replications with four ranges per replicate. Each replicate had 32 treatment combinations. Treatments were arranged as factorial combinations of tillage (hoe tillage and no-tillage), corn row spacing (50 and 75 cm), corn cultivar (OP corn ‘TZL-comp 4ci’ and hybrid corn ‘Oba super 1’), herbicide (glyphosate and no herbicide), and cover crop (velvetbean and no velvetbean). Tillage was accomplished by digging manually to a depth of approximately 30 cm using hoes. All plots designated for tillage were tilled on 10 June 1999 and 15 May 2000. All no-tillage treatments were slashed on 5 June in 1999 and 11 May in 2000, using machetes to minimize soil disturbance. Glyphosate was applied to all plots designated for this treatment at a rate of 1.0 kg a.e. (acid equivalent) ha–1 using a CP3 knapsack sprayer (Hardi International A/S, Taastrup, Denmark) on 29 June 1999 (19 d after tillage) and 7 June 2000 (22 d after tillage). The glyphosate was applied to the cogongrass at an average height of 68.42 cm. The sprayer was equipped with a blue polijet nozzle calibrated to deliver 250 L ha–1 at a pressure of 210 kPa. The two corn cultivars used in this study are recommended (MIP, 1996) for the forest/savanna transition zone of Nigeria. Plot size was 15 by 15 m. Since the trial was conducted on farmers' fields, large plot sizes were necessary to reduce errors and variability associated with farm conditions (Cady, 1991; Nokoe et al., 2000). All treatments were applied on the same plot in both years. Reduction of cogongrass to nondamaging levels usually requires repeated, consecutive, and integrated management over time (Chikoye et al., 2001; Chikoye and Ekeleme, 2003) because rhizomes with viable and dormant buds can sustain the regeneration of new cogongrass shoots (Soerjani, 1970; Eussen, 1981).

Corn was planted on 10 July 1999 and 1 July 2000 at a depth of 5 cm and at the same density per row to give a population of 80000 plants ha–1 for the 50-cm spacing and 53000 plants ha–1 for 75-cm row spacing. Corn seed was treated with Apron-Plus {10% metalaxyl [methyl N-(2,6-dimethylphenyl)-N-(methoxyacetyl)-DL-alaninate] + 6% carboxin (5,6-dihydro-2-methyl-N-phenyl-1,4-oxathiin-3-carboxamide) + 34% furathiocarb (2,3-dihydro-2,2-dimethyl-7-benzofuranyl 2,4-dimethyl-5-oxo-6-oxa-3-thia-2,4-diazadecanoate)} at the rate 10 g kg–1 of seed to protect against fungal infection, particularly downy mildew [Peronosclerospora sorghi (Weston and Uppal) Shaw]. Velvetbean [collected from the International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria], was planted at a depth of 5 cm into designated plots in rows alternate to each corn row 37 d and 31 d after corn sowing in 1999 and 2000, respectively. Velvetbean was planted at a population of approximately 53000 plants ha–1. The velvetbean seeds were soaked in 0.2% (w/v) benlate [methyl-1- (butylcarbamoyl) benzimidazol-2-yl carbamate] solution for 20 min and air-dried before planting to protect them against fungal diseases. All crops were planted by hand.

After planting, all the vegetation except velvetbean or corn in all plots was slashed using machetes on two occasions: at the time of applying basal fertilizer [N, P, and K; 2 wk after planting (WAP)] and urea (6 WAP). Slashing was used to remove weeds because it minimized soil disturbance that would have confounded tillage effects. Slashing twice, without applying other interventions, does not adequately control cogongrass, and usually results in corn yield losses (Chikoye et al., 2001). Basal fertilizer was applied at a recommended rate of 45 kg ha–1 for N, P2O5, and K2O. Urea at 45 kg N ha–1 was sidedressed on corn at 6 WAP. Mean monthly rainfall data were recorded near the experimental area (Fig. 1) .



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 1. Total monthly precipitation during the growing season (July to October) in 1999 and 2000 and 10-yr mean monthly precipitation (1988–1998).

 
All data except corn yield were collected from five fixed points that were equidistant (3 m apart) along two diagonals of each treatment plot, making 10 sampling points per plot. Corn leaf area index (LAI) per plot was determined nondestructively, with a LI-COR LAI–2000 canopy analyzer (LI–COR Inc., Lincoln, NE), at 50% corn silking (11 WAP) on 22 Sept. 1999 and 12 Sept. 2000. At this growth stage, the corn cultivars were expected to have reached maximum growth, and thus the impact of weed competition could be detected. One measurement above the canopy and five measurements below the canopy were made perpendicular to the corn row on two diagonal transects (Hicks and Lascano, 1995). The below-canopy measurements were made directly underneath the corn row at 20 cm above the ground and below the last corn leaf. Weeds above 20 cm were flattened to ground level before measurements were made. In velvetbean plots, LAI measurements were computed for both the cover crop and corn combined because, at the time of measurement, the cover crop had climbed the corn plants. Photosynthetically active radiation transmitted through the canopy (TPAR) was determined with a 1-m-long Decagon sunfleck ceptometer (Decagon Devices, Inc., Pullman, WA). Measurements were made under clear-sky conditions at solar noon (±2 h) at 50% corn silking, on the same day as LAI measurements. The incident photosynthetically active radiation (PAR) above the corn canopy was measured by taking three readings, 1 m above the ground, in the open without vegetation. The average of these readings was taken as the incoming PAR above the corn canopy. Measurement of TPAR below the corn canopy, 20 cm above the ground or weed canopy, was made by taking five readings perpendicular to the corn row on two fixed diagonal transects across each plot. The average from the replicated readings per plot represented the TPAR value for each treatment within a replicate.

Corn height was taken in each plot by measuring 10 plants at 50% corn silking. Cogongrass and other weeds were sampled from four quadrats measuring 0.25 m2 each at fixed points in each plot at corn harvest in 1999 and 2000. Shoots in each quadrat were clipped at ground level and cogongrass rhizomes excavated from a depth of 25 to 30 cm for biomass determination. This depth of excavation was necessary to assess the extent of cogongrass rhizome control because, at this depth, significant rhizome activity is possible, and beyond this depth, viability is doubtful. Samples were oven-dried at 80°C (Model OVE-300 Plus, Gallenkamp, Loughborough, UK) until constant mass was recorded with a digital balance (XD–4K B042809, Denver Instrument Co., Denver, CO). Corn was harvested on 29 Oct. 1999 and 24 Oct. 2000 from a net plot of 150 m2 at the center rows of each plot, excluding 1 m from the edge. This gave approximately 800 plants from the 16 middle rows in 75-cm row-spacing plots and 1200 plants from the middle 24 rows of 50-cm row-spacing plots. Grain yield was adjusted to moisture content of 12% with Tri-Grain moisture tester (Model 14998, Dickey-John Corp., Auburn, IL).

All the data were subjected to analysis of variance, using the PROC MIXED procedures in the Statistical Analysis Systems software (SAS Inst., 1995; Littell et al., 1996). Preliminary ANOVA with covariance analysis was conducted to remove the residual treatment effects of 1999 on 2000 plots since the same experimental plots were used in both years (data not shown). While years were significantly different, the results of main effects and interaction effects were similar. Therefore, the data were combined and means presented across years. All treatments were considered fixed in determining the expected mean squares and appropriate F tests in the analysis of variance. Replicates and years were considered as random factors. Treatment differences were separated using the probability of difference (PDIFF; SAS Inst., 1995) for sources of variation determined to be significant (P ≤ 0.05). Pairwise comparison between treatment least square means (LSMEANS) was considered different at P ≤ 0.05. Spearman's correlation analysis was performed to determine the relationship between weed biomass and LAI, TPAR, corn height, and corn grain yield (Steel and Torrie, 1980).


   RESULTS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Precipitation
Precipitation during the growing season (July–October) in 1999 and 2000 was generally similar or above the long-term (10-yr) average for this location, except in August and September in 1999 and October 2000 (Fig. 1). The distribution of precipitation was better in 2000 than in 1999.

Cogongrass Biomass
The main effects of tillage, herbicide, and cover cropping significantly (P = 0.0002 to 0.02) reduced cogongrass shoot biomass (Tables 1 and 2). Corn cultivars and row spacing did not differ significantly in their effect on cogongrass shoot biomass. There were interaction effects of tillage x herbicide and herbicide x cover crop combinations on shoot biomass (Table 2). Hoe tillage without glyphosate application reduced cogongrass shoot biomass by 75%, glyphosate application with no tillage reduced biomass by 98%, and hoe tillage combined with glyphosate application reduced shoot biomass by 90% compared with no tillage and no glyphosate (Fig. 2a) . Glyphosate combined with velvetbean reduced cogongrass shoot biomass greatly compared with glyphosate or velvetbean alone (Fig. 3a) . The control plots without velvetbean and glyphosate had the highest cogongrass shoot biomass. Hoe tillage combined with glyphosate and velvetbean reduced cogongrass shoot biomass more than the single effect of velvetbean or hoe tillage but not the combined effects of hoe tillage and glyphosate, glyphosate alone, or glyphosate plus velvetbean (Fig. 4a) .


View this table:
[in this window]
[in a new window]
  		Table 1. Effect of tillage, corn cultivar, row spacing, herbicide, and cover crop on biomass of cogongrass shoots (SDM), cogongrass rhizomes (RDM), cogongrass shoots + rhizomes (TDM), and other weeds (ODM) at corn harvest (means of 1999 and 2000 combined).

 

View this table:
[in this window]
[in a new window]
  		Table 2. Significance of tillage, cultivar, spacing, herbicide, cover crop, and their interaction on biomass of cogongrass shoots (SDW), cogongrass rhizomes (RDW), cogongrass shoots + rhizomes (TDW), and other weeds (ODW) at corn harvest (combined ANOVA of 1999 and 2000).

 


View larger version (29K):
[in this window]
[in a new window]
  		Fig. 2. Effect of tillage and herbicide on (a) cogongrass biomass and (b) corn grain yield at corn harvest (means of 1999 and 2000 combined). Control = no tillage and no herbicide, TH = tillage and herbicide, T = tillage only, and H = herbicide only (glyphosate). Bars indicate the standard error of the mean.

 


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3. Effect of herbicide and cover crop on (a) cogongrass biomass and (b) corn grain yield at corn harvest (means of 1999 and 2000 combined). Control = no glyphosate no cover crop, HC = herbicide and cover crop, C = cover crop only, and H = herbicide only. Bars indicate standard error of the mean.

 


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 4. Effect of tillage, herbicide, and cover crop on (a) cogongrass biomass and (b) corn grain yield at corn harvest (means of 1999 and 2000 combined). Control = no tillage, no herbicide, and no cover crop; C = cover crop only; T = tillage only; TH = tillage and herbicide; THC = tillage, herbicide, and cover crop; TC = tillage and cover crop; HC = herbicide and cover crop; and H = herbicide only. Bars indicate standard error of the mean.

 
Velvetbean significantly (P = 0.0409) reduced cogongrass rhizome biomass (Table 1 and 2). Significant interactions on rhizome biomass were found for tillage x herbicide (Table 2). Hoe tillage alone, glyphosate alone, and hoe tillage plus glyphosate reduced rhizome biomass by 43 to 65% compared with no tillage and no glyphosate (Fig. 2a). The effect of hoe tillage alone, hoe tillage plus glyphosate, and glyphosate alone on rhizomes was similar (Fig. 2a). Cover cropping with velvetbean contributed significantly in lowering rhizome biomass especially when combined with hoe tillage or glyphosate (Fig. 3a and 4a). Velvetbean alone, hoe tillage alone, glyphosate alone, hoe tillage plus glyphosate, hoe tillage plus velvetbean, glyphosate plus velvetbean, and hoe tillage plus glyphosate combined with velvetbean had similar effects on rhizome biomass, and all were better than the control with no tillage, no glyphosate, and no velvetbean (Fig. 4a).

Hoe tillage and velvetbean significantly (P < 0.05) reduced total cogongrass dry biomass (shoot + rhizome) compared with no tillage and no velvetbean (Table 2). Glyphosate, corn row spacing, and cultivar did not have any effect (P > 0.05) on cogongrass biomass (Tables 1 and 2). Significant (P < 0.05) interactions on total cogongrass biomass were found with tillage x herbicide and herbicide x cover crop (Table 2). Hoe tillage combined with glyphosate (29.5 ± 17.4 g m–2) reduced total cogongrass biomass by 58% compared with the no-tillage and no-glyphosate control. However, the effect of hoe tillage plus glyphosate was not significantly different from hoe tillage alone (23.8 ± 17.4 g m–2) and glyphosate alone (26.5 ± 10.1 g m–2). In the herbicide x cover crop interaction, glyphosate combined with velvetbean (27.0 ± 10.1 g m–2) had a significant effect on total cogongrass biomass compared with no glyphosate and no velvetbean (70.4 ± 9.9 g m–2). But the combined effect of glyphosate and velvetbean did not differ significantly from velvetbean alone (23.7 ± 17.0 g m–2) or glyphosate alone (29.1 ± 17.3 g m–2).

Biomass of Other Weeds (Excluding Cogongrass)
Across the 2 yr, the weed community (excluding cogongrass) was dominated by sedges (Cyperus spp., Kyllinga spp., and Mariscus alternifolius Vahl.), old world diamondflower (Oldenlandia coryombosa L.), life plant (Biophytum petersianum Klotzsch), and Platostoma africanum P. Beauv. These weeds occurred at densities of 79, 53, 38, and 23 plants m–2, respectively, and contributed 86% to the total weed community. Minor weed species that occurred at densities below 10 plants m–2 were siamweed [Chomolaena odorata (L.) King & Robinson], carry me seed (Phyllanthus amarus Schum. & Thonn), mock bluestem [Euclasta condylotricha (Hochst.Ex Steud.) Stapf], ricegrass paspalum (Paspalum polystachyum R. Br.), tropical crabgrass (Digitaria adscendens Kunth Heenr.), Stylochiton lancifolius Kotschy & Peyr., and tropical spiderwort (Commelina benghalensis L.). The shoot biomass of the dominant weed community was lower (P ≤ 0.05) in the treatments with narrow corn row spacing and in those with velvetbean (Table 1). There were significant cultivar x cover crop (P = 0.0459) and tillage x herbicide x cover crop (P = 0.04) interactions for biomass of other weeds (Table 2). All the other interactions were not were not significant at the 5% level (P > 0.0614). Both OP (36.4 ± 13.4 g m–2) and hybrid (33.7 ± 6.4 g m–2) corn cultivars, combined with velvetbean, reduced the biomass of other weeds by 56 and 71%, respectively, when compared with OP (82.6 ± 6.4 g m–2) and hybrid (114.1 ± 6.2 g m–2) cultivars without velvetbean. Hoe tillage, combined with glyphosate and cover crops, had the lowest biomass of other weeds (33.5 ± 12.4 g m–2) while plots with no tillage, no glyphosate, and no cover crops had the highest biomass (82.9 ± 12.4 g m–2). All three-way treatment combinations that included cover cropping had lower biomass than any other treatment, especially those that had glyphosate.

Transmitted Photosynthetically Active Radiation
The OP corn cultivar, narrow corn rows, glyphosate, and velvetbean significantly (P ≤ 0.05) reduced TPAR under the corn canopy compared with the hybrid cultivar, no glyphosate, wide row spacing, and no velvetbean treatments, respectively (Table 3 and 4). Significant interactions for TPAR were for tillage x herbicide only. No-tillage with glyphosate (22.8 ± 2.1%), tillage without glyphosate (23.3 ± 3.5%), and tillage with glyphosate (24.8 ± 2.1%) had lower TPAR than no-tillage without glyphosate (32.9 ± 2.1%).


View this table:
[in this window]
[in a new window]
  		Table 3. Effect of tillage, corn cultivar, row spacing, herbicide, and cover crop on transmitted photosynthetically active radiation (TPAR), leaf area index (LAI), corn height, and grain yield (means of 1999 and 2000 combined).

 

View this table:
[in this window]
[in a new window]
  		Table 4. Significance of tillage, cultivar, spacing, herbicide, cover crops, and their interaction on transmitted photosynthetically active radiation (TPAR), leaf area index (LAI), corn height, and corn grain yield (combined ANOVA of 1999 and 2000).

 
Corn Leaf Area Index
Tillage, spacing, and cover cropping were significant main effects on LAI (P < 0.05, Table 4). Hoe tillage increased LAI slightly (Table 3). Corn planted in narrow rows contributed to higher canopy LAI than wide rows, which was to be expected, considering the effect of this treatment on plant density, i.e., there were 80000 seeds ha–1 for 50-cm row spacing and 53000 ha–1 for the 75-cm row spacing (Table 3). The use of the cover crop increased the LAI because the cover crop contributed to the total leaf area (Table 3). Leaf area index was not influenced by cultivars or glyphosate application. There was no significant interaction or treatment combination effect found for LAI (Table 4). The use of narrow rows increased the LAI more than wider rows, either individually or combined with any other treatment.

Corn Height
Corn height responded significantly (P < 0.0014) to the effects of hoe tillage and cultivar only (Table 4). Hoe tillage and the OP corn cultivar had significantly taller corn than no-tillage and the hybrid cultivar, respectively (Table 3). No significant treatment interactions were found (Table 4). Increase in competitive ability of crop cultivars has been linked to plant height (Barnes et al., 1990).

Corn Grain Yield
All main effects had a significant effect on corn grain yield (Table 3), and there were no significant treatment interactions found for grain yield (Table 4; Fig. 2b, 3b, and 4b). Corn grain yield was significantly (P < 0.05) higher in plots without velvetbean than in plots with velvetbean, in the OP corn cultivar than in hybrid cultivar, with glyphosate than no glyphosate herbicide, in 50-cm corn row spacing than in 75-cm corn row spacing, and in hoe tillage treatments than in no-tillage treatments (Table 3). Tillage increased corn grain yield by 21%, narrow corn row spacing by 20%, and glyphosate by 21% compared with no-tillage, wide row spacing, and no-glyphosate treatments, respectively. The OP corn cultivar had 38% higher grain yields than the hybrid cultivar (Table 3).

Correlations between Corn and Weed Growth Parameters
Corn LAI was negatively correlated with cogongrass as follows: shoot biomass, r = –0.33 and P = 0.0016; rhizome biomass, r = –0.44 and P < 0.0001; and total biomass, r = –0.48 and P < 0.0001 (Table 5). Cogongrass shoot (r = 0.50, P < 0.0001), rhizome (r = 0.34, P = 0.0007), and total biomass (r = 0.45, P < 0.0001) were positively correlated with TPAR (Table 5). The biomass of other weeds was negatively correlated with corn LAI (r = –0.62, P = 0.0001, Table 5). Cogongrass shoot and rhizome biomass were negatively correlated with corn height (r = –0.33 to –0.23, P < 0.05). There was a negative correlation between corn grain yield and cogongrass shoot biomass (r = –0.21, P = 0.0338), rhizome biomass (r = –0.18, P = 0.0754), and total cogongrass biomass (r = –0.23, P = 0.0244). Corn height was positively correlated with canopy LAI and corn grain yield and negatively correlated with TPAR.


View this table:
[in this window]
[in a new window]
  		Table 5. Spearman correlation coefficients between cogongrass shoot biomass (SDM), rhizome biomass (RDM), cogongrass total biomass (TDM), biomass of other weeds (ODM), and corn height, leaf area index (LAI), transmitted photosynthetically active radiation (TPAR), and corn grain yield (means of 1999 and 2000 combined).

 

   DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Results from this study showed that tillage, glyphosate, and cover cropping with velvetbean had a significant impact on cogongrass biomass. These results are in agreement with previous studies, which showed the superiority of these three control options for cogongrass management in West Africa (Ivens, 1980; Akobundu et al., 2000; Chikoye et al., 2002). Integrating the two or three control options gave better control of cogongrass than the main effects of each option. However, the effect of glyphosate alone in reducing biomass was comparable to integrating two or three control options. Although herbicides are an important tool in the management of cogongrass, farmers consider that they are more expensive than using family labor (Chikoye et al., 2001). In areas where labor availability is limiting, farmers have adopted the use of chemical control (Ellis-Jones et al., 2003).

The effect of combining tillage with velvetbean was also similar to combing glyphosate with velvetbean, but velvetbean plus glyphosate was better than tillage plus glyphosate. Combining hoe tillage with velvetbean had a greater effect in reducing rhizome biomass than any other combination of options. The effect of velvetbean on cogongrass has been attributed to shading, which reduces the carbohydrate content and growth vigor of the weed (Moosavi-nia and Dore, 1979; Chikoye and Ekeleme, 2001). Moosavi-nia and Dore (1979) reported that the activity of glyphosate is better on cogongrass that has been growing under shaded conditions. This may explain why the interactive effect of glyphosate and velvetbean gave better control of cogongrass. Better control of cogongrass can be attributed to an increase in LAI resulting from canopy closure in narrow rows, better rainfall distribution pattern during the growing period, and the integration of velvetbean.

This study showed that effective control of cogongrass results in increased dominance of sedges and annual broadleaved weeds. Other studies have also reported similar observations in the savanna of West Africa (Anoka et al., 1991; Udensi et al., 1999). Sedges, notably Cyperus spp., are among the world's 10 most infamous weeds because they cause significant losses in many crops (Holm et al., 1977). Sedges are difficult to control in small-scale farming systems that depend on hoe weeding because they possess underground tubers and rhizomes. The dominant annual broadleaved weeds (old world diamondflower, life plant, and P. africanum) are not regarded as serious weeds because one hoe weeding can control them.

Better control of weeds other than cogongrass was achieved through the use of narrow corn row spacing, cover cropping with velvetbean, and an integrated approach comprising corn cultivar and cover cropping with velvetbean. Narrow corn row spacing and the use of velvetbean may have resulted in earlier canopy closure, and this effect resulted in more shading of other weeds as indicated by low TPAR in these treatments. These observations are supported by the negative correlation between LAI and corn height and the biomass of other weeds. Our observations are consistent with the work of Murphy et al. (1996), who showed that narrow crop rows reduced the biomass of late-emerging weeds. Weeding performed 2 and 6 WAP in our study may have delayed emergence of other weeds. Udensi et al. (1999) found that annual weeds increased after effective control of cogongrass with glyphosate, unless chemical control was integrated with the use of cover crops in Nigeria. When combined with velvetbean, the OP corn cultivar may have contributed to more shading of other weeds because of its greater height and LAI, which reduced the TPAR reaching the annuals under the corn canopy. The OP cultivar used in this study has a greater height (195 to 236 cm) than the hybrid cultivar (214 vs. 183 cm, SE = 7 cm, respectively). Seavers and Wright (1999) reported that higher LAI and height were among attributes that gave crops more competitive ability against weeds.

Hoe tillage, glyphosate, and narrow corn row spacing had a positive effect on corn grain yield because they reduced the biomass of cogongrass (Fig. 2b, 3b, and 4b). Previous studies have shown that hoe tillage and the use of glyphosate had a negative impact on cogongrass biomass, which results in higher crop yields (Chikoye et al., 2002). However, the high cost of labor often prevents the effective use of hoe tillage by smallholder farmers in West Africa. The positive benefit from narrow row spacing for weed suppression and increased corn yield obtained in our study is consistent with previous studies (Bullock et al., 1988; Murphy et al., 1996; Cox and Cherney, 2001). This study has shown that the OP cultivar has a higher yield potential than the hybrid (1861 vs. 1351 kg ha–1, SE = 130 kg ha–1, respectively), an observation that agrees with previous studies (MIP, 1996).

We recorded lower corn grain yields where velvetbean was used to suppress cogongrass (Fig. 2b, 3b, and 4b). Low corn grain yields in this treatment could not have been due to weed competition because the cover crop treatment had the lowest total weed biomass. Since velvetbean was planted 1 wk before the recommended planting time (Versteeg and Koudokpon, 1990), it is likely that the cover crop competed aggressively with corn. Chikoye et al. (2002) have reported lower corn yields where velvetbean was intercropped with corn or cassava. To minimize the competitive effects of velvetbean on corn, it is suggested that a sole cover crop be grown for the entire rainy season, and then followed by a food crop in the subsequent year, where land is not a limiting factor (Chikoye et al., 2002). Other options for minimizing competition include lower populations of cover crops or the use of less aggressive cover crops.


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
The major findings of this study were (i) integrating tillage, glyphosate, and cover cropping with velvetbean gave better control of cogongrass than the main effects of each option. However, if only one control option can be adopted, glyphosate is the most effective in the short run, but when considering the control of subsequent regrowth, an integrated approach, especially with cover crops, will be the best option. (ii) Better control of sedges and annual broadleaved weeds was achieved through the use of narrow corn row spacing and cover cropping with velvetbean. (iii) Tillage, narrow corn row spacing, the OP cultivar, and the use of glyphosate had a positive effect on corn grain yield. Cultivar selection and row spacing are easy management options, and integrated with one or more other tools (e.g., herbicides) over time will be effective in reducing cogongrass to noncompetitive levels in corn.


   ACKNOWLEDGMENTS
 
The authors appreciate the technical assistance provided by John Ogazie and Kayode Sanyaolu and land provided by the Ijaye community in Ibadan.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
This article is published with approval from IITA as contribution no. IITA/05/JA/37.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 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 Similar articles in ISI Web of Science
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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chikoye, D.
Right arrow Articles by Ogunyemi, S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Chikoye, D.
Right arrow Articles by Ogunyemi, S.
Agricola
Right arrow Articles by Chikoye, D.
Right arrow Articles by Ogunyemi, S.
Related Collections
Right arrow Weed Management
Right arrow Agricultural Pesticides
Right arrow Cover Crops

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