BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. Effect of Glyphosate Application Timing and Row Spacing on Corn (Zea mays) and Soybean (Glycine max) Yields Author(s): CALEB D. DALLEY, JAMES J. KELLS, KAREN A. RENNER Source: Weed Technology, 18(1):165-176. 2004. Published By: Weed Science Society of America DOI: http://dx.doi.org/10.1614/02-150A URL: http://www.bioone.org/doi/full/10.1614/02-150A BioOne (www.bioone.org ) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use . Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.
Transcript
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, researchlibraries, and research funders in the common goal of maximizing access to critical research.
Effect of Glyphosate Application Timing and Row Spacing on Corn (Zea mays)and Soybean (Glycine max) YieldsAuthor(s): CALEB D. DALLEY, JAMES J. KELLS, KAREN A. RENNERSource: Weed Technology, 18(1):165-176. 2004.Published By: Weed Science Society of AmericaDOI: http://dx.doi.org/10.1614/02-150AURL: http://www.bioone.org/doi/full/10.1614/02-150A
BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, andenvironmental sciences. BioOne provides a sustainable online platform for over 170 journals and books publishedby nonprofit societies, associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance ofBioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiriesor rights and permissions requests should be directed to the individual publisher as copyright holder.
Effect of Glyphosate Application Timing and Row Spacing on Corn (Zea mays) andSoybean (Glycine max) Yields1
CALEB D. DALLEY, JAMES J. KELLS, and KAREN A. RENNER2
Abstract: Corn and soybean were planted in narrow and wide row spacings to determine the effectof glyphosate application timing and row spacing on crop yield. Glyphosate was applied whenaverage weed canopy height reached 5, 10, 15, 23, and 30 cm. Weeds present in these studies includedvelvetleaf, redroot pigweed, common ragweed, common lambsquarters, jimsonweed, barnyardgrass,fall panicum, giant foxtail, yellow foxtail, green foxtail, and eastern black nightshade. Under highlycompetitive growing conditions (below normal rainfall and high weed density), corn yield was firstreduced when weeds reached 10 and 15 cm in height with corn planted in 38- and 76-cm rows,respectively. Under similar conditions, soybean yield was first reduced when weeds reached 15 and23 cm with soybean planted in 19- and 38-cm rows, respectively. Yield losses occurred only in theuntreated control when soybean was planted in 76-cm rows. When growing conditions were lesscompetitive (adequate rainfall and lower weed density), yield losses occurred only when weedsreached 30 cm or more in corn and only in the untreated control in soybean. Corn and soybeanyields were higher when planted in narrow rows in three of 4 yr but were more susceptible to early-season weed interference than corn and soybean in wide rows. Corn yield was affected more byweed interference than was soybean yield. The product of weed height by weed density, as theindependent variable, resulted in the best linear fit for both corn and soybean yields. High weeddensities increase the risk of yield loss and must be considered when determining the appropriatetiming for total postemergence herbicide applications such as glyphosate. Sequential glyphosate ap-plications in corn did not increase yield.Nomenclature: Glyphosate; barnyardgrass, Echinochloa crus-galli (L.) Beauv. #3 ECHCG; commonlambsquarters, Chenopodium album L. # CHEAL; common ragweed, Ambrosia artemisiifolia L. #AMBEL; eastern black nightshade, Solanum ptycanthum Dun. # SOLPT; fall panicum, Panicumdichotomilflorum Michx. # PANDI; giant foxtail, Setaria faberi Herrm. # SETFA; green foxtail,Setaria viridis (L.) Beauv. # SETVI; jimsonweed, Datura stramonium L. # DATST; redroot pigweed,Amaranthus retroflexus L. # AMARE; velvetleaf, Abutilon theophroasti Medik. # ABUTH; yellowfoxtail, Setaria glauca (L.) Beauv. # SETLU; corn, Zea mays L. ‘DK 493RR’; soybean, Glycinemax (L.) Merr. ‘Pioneer 92B71’.Additional index words: Application timing, competitive load, narrow-row corn, narrow-row soy-bean, weed competition, weed interference.Abbreviations: DAE, days after crop emergence; GDDAE, growing degree days after emergence;POST, postemergence; SAS, Statistical Analysis Systems.
INTRODUCTION
Weeds compete with crops for soil moisture, sunlight,and nutrients, and when any of these resources becomelimited, crop and weed growth can be constrained. The
1 Received for publication September 22, 2002, and in revised form July30, 2003.
2 Former Graduate Research Assistant, Professor, and Professor, Departmentof Crop and Soil Sciences, Michigan State University, East Lansing, MI48824. Corresponding author’s E-mail: [email protected].
3 Letters following this symbol are a WSSA-approved computer code fromComposite List of Weeds, Revised 1989. Available only on computer diskfrom WSSA, 810 East 10th Street, Lawrence, KS 66044-8897.
period of time in which weed control efforts affect cropyield is known as the critical period for weed controland is defined by two terms, the weedy period (how longweeds can remain within a crop before yield losses beginto occur) and the weed-free period (how long weed con-trol efforts must be maintained to prevent yield loss)(Weaver et al. 1992). Hall et al. (1992) reported that thebeginning of the critical period for weed control in cornvaried according to weed densities and environmentalconditions more than the end of the critical period ofweed control. The introduction of herbicide-resistant
DALLEY ET AL.: EFFECT OF GLYPHOSATE APPLICATION TIMING AND ROW SPACING
166 Volume 18, Issue 1 (January–March) 2004
crops has not only greatly enabled the study of the crit-ical period for weed control, through the use of nonse-lective herbicides such as glyphosate, but also created anecessity for its study.
The first genetically transformed herbicide-resistantcrop, glyphosate-resistant soybean (Padgette et al. 1995),was first grown commercially in 1996. Today, glyphos-ate-resistant corn hybrids, cotton (Gossypium hirsutumL.), and canola (Brassica napus L.) cultivars also arecommercially available. With glyphosate-resistant crops,most annual weeds and some perennial weeds can becontrolled with a postemergence (POST) glyphosate ap-plication. Weed control is achieved with increased flex-ibility when using glyphosate-resistant crops because ofglyphosate’s effectiveness on controlling weeds regard-less of size (Ateh and Harvey 1999; Krausz et al. 1996;Tharp and Kells 1999; Tharp et al. 1999), although high-er application rates may be needed to control largerweeds (Krausz et al. 1996). The most important factorin correctly timing glyphosate applications is avoidanceof yield loss due to early-season weed interference(Krausz et al. 2001). Greater understanding of the criticalperiod of weed control can help growers make betterdecisions regarding the timing of herbicide application,especially when planting herbicide-resistant crops.
The proper glyphosate application timing needed toachieve satisfactory weed control without yield loss dueto competition is not well understood. Many studies havebeen conducted to examine the effect of weed removaltiming on crop yield; however, in these studies weedremoval timing has been based on a variety of factors,including weed height (Knake and Slife 1969; Krausz etal. 2001), crop growth stage (Hall et al. 1992; Van Ges-sel et al. 2000), and days or weeks after crop emergence(Horn and Burnside 1985). Tharp and Kells (1999) stat-ed that corn yield reductions, when glyphosate was ap-plied late-POST, may be due to early-season weed inter-ference. Carey and Kells (1995) concluded that delayingPOST herbicide applications increased the probability ofyield reductions in corn. Horn and Burnside (1985)found that soybean yield was not reduced when weedswere removed 4 wk after emergence but that whenweeds remained in the crop for 6 wk, yields were re-duced in five of eight experiments. Van Gessel et al.(2000) stated that weed control and yield were greatestwhen glyphosate was applied to soybean at the one- tothree-trifoliate stage (18 to 28 d after planting). Krauszet al. (2001) found that soybean could tolerate weed in-terference until weeds reached 30 cm in height (V5 leafstage; 6 to 7 wk after planting) without sacrificing yield.
Use of a single factor, such as weed height, cropheight, or crop growth stage, as a basis for applicationtiming is common when conducting research on the ef-fects of herbicide application timing on weed interfer-ence. However, other factors such as environmental con-ditions, weed density, and species composition may alsoplay an important role in determining when yield lossesfirst occur. Also, it is not clear which parameter, weedheight, crop growth stage, days after crop emergence(DAE) should be used as a basis for timing herbicideapplications.
Interest in the use of narrow-row crop production hasexisted for many years (Wiggins 1939). Interest in nar-row rows stems from yield advantages associated withplanting in narrow rows, which are generally attributedto more efficient use of sunlight, especially early in thegrowing season (Bullock et al. 1998). Narrow-row soy-bean has an increased number of pods per plant, branch-ing, plant survival, and harvest index (Board and Har-ville 1993, 1996; Egli 1994; Ethredge et al. 1989; Ikeda1992). Nelson and Renner (1999) reported a $44 to $64/ha increase in gross margin for narrow-row (19 cm) soy-bean compared with wide-row (76 cm) soybean. Forcorn, the advantages of narrow row widths are more var-iable. Bullock et al. (1988) found that corn planted atequidistant spacing in 38-cm rows consistently yieldedgreater than corn planted in 76-cm rows at the samepopulation, whereas Westgate et al. (1997) reported noyield gains for corn planted in narrow rows. Widdicombeand Thelen (2002) reported that the advantages of usingnarrow row spacings for corn were greater at central andnorthern locations than at southern locations in Michi-gan. This report is consistent with the findings of Pasz-kiewicz (1997), who suggested that the greatest yieldresponse to narrow rows occurred north of the 448N lat-itude.
Herbicide-resistant crops allow the study of POSTweed control in different crops using identical weed con-trol practices. A greater understanding of the yield ben-efits of narrow row spacings and the effects of row spac-ing on weed interference is needed to refine weed controlrecommendations for herbicide-resistant crops. The ob-jectives of this study were (1) to determine the effect ofrow spacing and glyphosate application timing on cornand soybean yields and (2) to compare the relative tol-erance of soybean and corn to early-season weed inter-ference.
MATERIALS AND METHODS
At East Lansing, MI, experiments were conducted onthe same fields in 1998 and 2000, with an adjacent field
WEED TECHNOLOGY
Volume 18, Issue 1 (January–March) 2004 167
Tab
le1.
Trea
tmen
tda
tes
and
crop
and
wee
dhe
ight
sat
Eas
tL
ansi
ng,
MI.
Fact
ors
1998
5a10
1523
30
1999
510
1523
30
2000
510
1523
30
2001
510
1523
30
DA
Eb,
c
GD
DA
EC
orn
heig
ht(c
m)
Cor
ngr
owth
stag
eSo
ybea
nhe
ight
(cm
)So
ybea
ngr
owth
stag
e
23 409 20 V3
— —
26 491 25 V4
13 V3
30 618 30 V4 15 V3
35 786 58 V6 23 V5
38 877 64 V7
— —
15 313 13 V3 8
V1
19 458 18 V4 10 V2
21 538 28 V5 13 V2
25 640 33 V6 18 V3
31 764 53 V7 23 V5
22 529 35 V6 13 V2
23 553 36 V6 13 V2
26 642 46 V7 15 V3
29 705 56 V8 18 V4
34 833 71 V9 25 V5
21 366 10 V3 8
V2
25 495 18 V4 10 V2
27 561 23 V5 15 V3
30 628 25 V5 18 V4
33 703 36 V6 20 V4
aG
lyph
osat
eap
plic
atio
ntim
ing
was
acco
rdin
gto
wee
dhe
ight
(cm
).G
lyph
osat
ew
asap
plie
dat
0.84
kgae
/ha.
Sequ
entia
lgl
ypho
sate
appl
icat
ions
(0.4
2kg
ae/h
a)w
ere
mad
ein
corn
afte
rin
itial
appl
icat
ions
at5-
,10
-,an
d15
-cm
heig
hts
ofw
eeds
,an
dw
eeds
reac
hed
30cm
inhe
ight
inth
eun
trea
ted
chec
k.b
Abb
revi
atio
ns:
DA
E,
days
afte
rcr
opem
erge
nce;
GD
DA
E,
grow
ing
degr
eeda
ysaf
ter
crop
emer
genc
e.c
Cor
nan
dso
ybea
nw
ere
plan
ted
onM
ay19
,19
98,
May
14,
1999
,M
ay25
,20
00,
and
May
14,
2001
.
used in 1999 and 2001. The soil type in 1998 and 2000was Capac sandy loam (fine-loamy, mixed mesic AericOchraqualfs) with 56% sand, 30% silt, and 14% clay,pH 6.6, and 2.9% organic matter. In 1999 and 2001 thesoil was a Capac sandy clay loam (fine-loamy, mixedmesic Aeric Ochraqualfs) with 52% sand, 28% silt, and20% clay, pH 6.5, and 3.0% organic matter.
At Clarksville, MI, experiments were conducted in1998, 1999, and 2001 on three different fields. The soiltype in 1998 was a mix of Dryden sandy loam (coarse-loamy, mixed, mesic Oxyaquic Hapludalfs) and Lapeersandy loam (coarse-loamy, mixed, mesic Typic Haplu-dalfs) with 75% sand, 17% silt, and 8% clay, pH 6.7,and 1.8% organic matter. In 1999 and 2001 the soil typewas Lapeer loam (coarse-loamy, mixed, mesic Typic Ha-pludalfs) with 47% sand, 36% silt, and 17% clay, pH6.8, and 2.0% organic matter.
Fields were fall chisel plowed followed by a field cul-tivator in the spring for seedbed preparation. Beforeplanting, corn plots were fertilized with granular urea(46-0-0) at 300 kg/ha using a broadcast applicator andincorporated using a field cultivator. Soil tests showedthat adequate levels of phosphorous and potassium werepresent; therefore, no additional fertilizers were applied.
A glyphosate-resistant full-season corn hybrid ‘DK493RR’4 and an indeterminate group-II soybean cultivar‘92B71’5 were planted at 77,000 and 422,000 seeds/ha,respectively. Corn and soybean were planted using a cus-tomized toolbar with John Deere planter units. Corn wasplanted in two-row spacings (38- and 76-cm row widths)and soybean was planted in three-row spacings (19-, 38-,and 76-cm row widths). Plant populations remained con-stant across the different row spacings of corn and soy-bean. At East Lansing, corn and soybean were plantedon May 19, 1998, May 14, 1999, May 25, 2000, andMay 14, 2001. At Clarksville, corn and soybean wereplanted on May 19, 1998, and May 17, 1999, and cornwas planted on May 4, 2001. Plot size was 3 m wide by10.7 m long for each experiment.
Weeds were controlled using glyphosate6 at 0.84 kgae/ha plus 2% (w/w) ammonium sulfate. Weeds werecontrolled when the average weed canopy height reached5, 10, 15, 23, and 30 cm, although not all treatmenttimings were included in all experiments (Tables 1 and2). Weed-free and untreated checks were included for
4 DeKalb Genetics Corp., Monsanto County, 800 North Lindbergh Boule-vard, St. Louis, MO 63167.
5 Pioneer Hi-Bred International, 400 Locust Street, Suite 800, Des Moines,IA 50306.
6 Roundup Ultra, Monsanto County, 800 North Lindbergh Boulevard, St.Louis, MO 63167.
DALLEY ET AL.: EFFECT OF GLYPHOSATE APPLICATION TIMING AND ROW SPACING
168 Volume 18, Issue 1 (January–March) 2004
Table 2. Treatment dates and crop and weed heights at Clarksville, MI.
a Weed height. Glyphosate was applied at 0.84 kg ae/ha.b Abbreviations: DAE, days after crop emergence; GDDAE, growing de-
gree days after crop emergence.c Corn and soybean were planted on May 19, 1998, and May 17, 1999.
Corn was planted on May 4, 2001.
each crop species and each row spacing. Weed controlin weed-free plots was accomplished through preemer-gence application of S-metolachlor at 0.57 kg ai/ha plusatrazine at 0.45 kg ai/ha7 in corn and alachlor at 0.91 kgai/ha8 in soybean followed by POST glyphosate appli-cations or hand weeding as needed in both crops. Allherbicides were applied using a tractor-mounted com-pressed-air sprayer calibrated to deliver 187 L/ha at 207kPa using XR8003 flat-fan nozzles.9
Experimental Design. At East Lansing, the experimen-tal design was a split–split plot each year. The main plotwas crop species (corn or soybean), the subplot was rowspacing (19-, 38-, and 76-cm widths), and the sub-sub-plot was time of glyphosate application (Table 1). A 3-m border of corn and a 3-m border of soybean separatedeach main plot of corn and soybean to minimize the riskof border effects. In 1998, glyphosate was applied insoybean plots when weeds were 10, 15, and 23 cm inheight and in corn plots when weeds were 5, 10, 15, 23,and 30 cm in height. Glyphosate was applied whenweeds were 5, 10, 15, 23, and 30 cm in height in cornand soybean in 1999, 2000, and 2001. In addition, atEast Lansing, sequential glyphosate applications weremade in corn in each year of this study. The first gly-phosate (0.84 kg/ha) application was made when weedcanopy height reached 5, 10, and 15 cm, which was fol-lowed by a second glyphosate application (0.42 kg/ha)when weeds in the untreated control reached 30 cm. Allother timings received only a single glyphosate appli-cation. These treatment timings were compared withweed-free and untreated checks (Table 1).
7 Bicep Lite II Magnum, a mixture of 399 g S-metolachlor and 320 g at-razine per liter, Syngenta Crop Protection, Inc., P.O. Box 18300, Greensboro,NC 27419.
8 Lasso, Monsanto Co., 800 North Lindbergh Boulevard, St. Louis, MO63167.
9 TeeJet XR 8003, Spraying Systems Co., North Avenue, Wheaton, IL60188.
At Clarksville, smaller experiments were establishedto support the larger experiments conducted at East Lan-sing (Table 2). In 1998 and 1999, corn and soybean wereplanted in a split–split-plot design. Crop species (corn orsoybean) was the main plot, row spacing (38- and 76-cm row widths for corn, and 19-, 38-, and 76-cm rowwidths for soybean) was the subplot, and timing of her-bicide application was the sub-subplot. Glyphosate, at0.84 kg/ha, was applied when weeds were 15 cm inheight, which was compared with weed-free and untreat-ed checks. As at East Lansing, a 3-m border of corn anda 3-m border of soybean separated each whole plot ofcorn and soybean. In 2001, corn was planted in a split-plot design. The main plot was row spacing (38- and 76-cm row widths), and the subplot was time of herbicideapplication. Glyphosate, at 0.84 kg/ha, was applied whenweeds reached an average canopy height of 5, 10, 15,23, and 30 cm (Table 2).
Weed Density. Weed density was measured to quantifythe presence and density of the native weed populationsin each experiment. Weed densities, by species, weremeasured in the untreated plots using three 14- by 75-cm quadrats randomly placed parallel to the middle ofthe crop rows. Quadrats were marked with flags so thatdensity measurements would come from the same loca-tions within each plot each time they were measured.Densities were measured when weeds were 5, 10, and15 cm in height (Table 3). Competitive loads were cal-culated by summing the product of weed density andcompetitive index for each weed species.
Crop Yields. Corn and soybean were harvested at ma-turity. Corn was harvested using a plot combine with afive-row header designed to harvest corn planted in ei-ther 38- or 76-cm rows. The center two rows were har-vested from corn in 76-cm row widths (an area of 16m2), and the five middle rows were harvested from cornin 38-cm row widths (an area of 20 m2). Soybean washarvested using a plot combine with a 1.5-m header sothat the center two rows were harvested from soybeanin 76-cm rows, four rows were harvested from soybeanin 38-cm rows, and eight rows were harvested from soy-bean in 19-cm rows (an area of 16 m2 for each). Yieldand moisture were measured and recorded with a datalogger on the plot combine. Corn yield was adjusted to15.5% moisture, and soybean yield was adjusted to 13%moisture. Yields were calculated according to area har-vested. To compare the effects of weed interference oncorn and soybean, yields were converted to percentageof the weed-free check when these comparisons were
WEED TECHNOLOGY
Volume 18, Issue 1 (January–March) 2004 169
Table 3. Weed densities and competitive load values at East Lansing and Clarksville.a,b
LocationYear
Weed densities
ABUTH ANGR AMBEL AMARE CHEAL SOLPT DATST Total
CLc
Corn Soybean
weeds/m2
ELAN1998199920002001
9222115
181,011
10194
0000
29973925
42414012035
000
55
0005
5631,270
180319
4,7643,0941,414
834
2,2631,559
694491
CHES199819992001
000
00
80
010
281641
61391760
000
000
89407881
7223,9847,860
3561,6383,244
a Reported weed densities were measured when weed canopy height was 15 cm.b Abbreviations: ELAN, East Lansing; CHES, Clarksville Horticulture Experiment Station; ABUTH, velvetleaf; ANGR, annual grass (including giant foxtail,
green foxtail, yellow foxtail, barnyardgrass, and fall panicum); AMBEL, common ragweed; AMARE, redroot pigweed; CHEAL, common lambsquarters; SOLPT,eastern black nightshade; DATST, jimsonweed.
c CL 5 competitive load, which is derived by summing the products of weed densities and competitive indices (CI) for each weed. CIs for corn are ABUTH5 4.2, ANGR (giant foxtail) 5 1.2, AMBEL 5 10, AMARE 5 4.0, CHEAL 5 10. SOLPT 5 2.0, and DATST 5 4.0. CIs for soybean are ABUTH 5 4.8,ANGR (giant foxtail) 5 0.5, AMBEL 5 10, AMARE 5 4.0, CHEAL 5 4.0, SOLPT 5 2.0, and DATST 5 4.0.
Table 4. Monthly precipitation recorded at East Lansing, MI, and at Clarksville, MI.a
Month
East Lansing
1998 1999 2000 2001 Normalb
Clarksville
1998 1999 2001 Normalc
mm
MayJuneJulyAugust
49555783
4242
10344
135809386
145852441
69897679
39344155
6710210079
1366723
103
74976092
Totald 244 231 394 254 313 168 348 329 322
a Rainfall was recorded at the Michigan State University Department of Horticulture Teaching and Research Center in East Lansing and at the ClarksvilleHorticulture Experiment Station.
b Average monthly precipitation measured at East Lansing from 1951 to 1980.c Average monthly precipitation measured at Ionia from 1951 to 1979.d Total rainfall from May 1 to August 31.
made. In 1998, yield losses occurred in the weed-freecheck likely due to weekly light measurements taken inthese plots. Therefore, comparisons of yield results be-tween corn and soybean were not included for 1998.
Statistical Analysis. All data were analyzed using Sta-tistical Analysis Systems (SAS) 6.12 and 8.00 soft-ware.10 The PROC GLM procedure in SAS was used foranalysis of variance, and appropriate F-tests were run todetermine significance. Regression analyses of yield andthe application timing parameters, weed height, cropheight, crop growth stage, DAE, and growing degreedays after emergence (GDDAE), and the products ofthese parameters with weed density and competitive loadwere performed separately for each row spacing of cornand soybean. These analyses were performed to deter-mine which application timing parameter best predicted
10 SAS Institute Inc., 100 SAS Campus Drive, Cary, NC 27513.
the effect of glyphosate application timing on corn andsoybean yields.
RESULTS AND DISCUSSION
Growing Conditions. The diverse environmental con-ditions and weed densities that occurred during the yearsin which this research was conducted enabled studyingthe effects of weed interference on corn and soybeanunder different growing conditions. At East Lansing,weed density (Table 3) was greatest in 1999, and lowestin 2000. At East Lansing, rainfall (Table 4) was belownormal in May and June of both 1998 and 1999. AtClarksville, rainfall was below normal in May and Juneof 1998, and in May of 1999. In 2000, rainfall was abovenormal (especially in May) at East Lansing. Rainfall inthe 2001 growing season was variable at both locations.Rainfall in May was two times greater than normal,
DALLEY ET AL.: EFFECT OF GLYPHOSATE APPLICATION TIMING AND ROW SPACING
170 Volume 18, Issue 1 (January–March) 2004
Figure 1. Effect of glyphosate application timing on corn yield at East Lansingin 1998, 1999, and 2000 and combined from Clarksville and East Lansing in2001. Yields from 2000 and 2001 were averaged across row spacings. WF,weed free; UNT, untreated. Yield means with different lower-case letters aresignificantly different (P , 0.05) comparing treatment timings within rowspacing. Yield means with different upper-case letters are significantly differ-ent (P , 0.05) comparing row spacing within treatment timings.
which was followed by drought conditions beginning inthe latter part of June and running through the earlierpart of August.
Corn Yield. At East Lansing in both 1998 and 1999there was a significant interaction between time of gly-phosate application and row spacing for corn yield.Therefore, the data are presented showing the interactionwithin each row spacing. The time at which corn yieldwas reduced compared with the weed-free check variedby growing season. In 1999, under highly competitivegrowing conditions, yield losses occurred earlier than in2000 and 2001, when growing conditions were less com-petitive. Corn planted in narrow rows yielded more butwas more susceptible to weed interference than corn inwide rows (Figure 1).
In 1998, at no time was corn yield reduced comparedwith the weed-free check, except in untreated corn.When corn was planted in 38-cm rows, yield was great-
est when weeds were controlled at 5 cm in height andyield was reduced by weed interference when weedsreached 15 cm or more (V4 growth stage) (Figure 1).Corn yield also was reduced in the weed-free check com-pared with yield when weeds were controlled at 5 cm.Yield loss in the weed-free check was most likely relatedto disturbance caused by weekly light interception mea-surements taken in these plots. In later years, separateweed-free plots were included for taking these measure-ments. When corn was planted in 76-cm rows, yield wasreduced when weeds reached 30 cm or more (V7 growthstage) compared with yield when weeds were controlledat 10 cm in height. Yield was similar in all other treat-ment timings. Corn planted in 38-cm rows yielded morethan corn planted in 76-cm rows only at the earliest gly-phosate application timing (5-cm weed height) and in theuntreated check (Figure 1).
In 1999, when corn was planted in 38-cm rows, cornyield was reduced compared with the weed-free checkwhen weeds reached 10 cm or more (V4 growth stage),and when corn was planted in 76-cm rows, corn yieldreduced when weeds reached 15 cm or more (V5 growthstage). Corn yield was greater when corn was planted in38-cm rows compared with corn in 76-cm rows if weedswere controlled before reaching 10 cm in height (Figure1).
Under lower levels of weed interference in 2000, therewas no significant interaction between row spacing andtime of glyphosate application; however, differences be-tween time of glyphosate application and differences be-tween row spacings were both significant. In 2000, cornyield was reduced compared with the weed-free checkwhen weeds were allowed to reach 30 cm or more (V9growth stage) (Figure 1). Corn planted in narrow rowsyielded 6% greater than corn planted in wide rows, av-eraged across all treatment timings (data not shown).
Corn responded differently in 2001 compared withthose in any of the previous years because of abnormalrainfall conditions. May was more than twice as wet asnormal (at East Lansing, 145 mm of rainfall in May2001; normal rainfall is 69 mm) (Table 4). This wasfollowed by a drought period lasting from the end ofJune through the first half of August. From June 23through August 15 (53 d) only 29 mm of rain fell at EastLansing. Normal rainfall during this time period is 141mm. High rainfall, combined with moderate weed den-sities, resulted in less-competitive growing conditionsearly in the season. The drought conditions later in theseason, during pollination, when corn is most susceptibleto yield loss from moisture stress (Classen and Shaw
WEED TECHNOLOGY
Volume 18, Issue 1 (January–March) 2004 171
Figure 2. Effect of glyphosate application timing on corn yield at Clarksvillein 1998 and 1999, averaged across row spacings. WF, weed free; UNT, un-treated. Yield means with different lower-case letters are significantly different(P , 0.05).
1970), resulted in yield losses not entirely associatedwith weed interference. First, there was no significantyield increase associated with planting corn in narrowrows (data not shown). Porter et al. (1997) also foundthat climatic conditions that limited corn yield also lim-ited the advantage gained by planting corn in narrowrows. Because of the drought conditions, the photosyn-thetic advantage of planting in narrow rows was negatedby the lack of soil moisture, which limited crop produc-tion. Because there was no significant interaction be-tween time of glyphosate application and row spacingand no significant difference between row spacings, cornyield was averaged across row spacings. Corn yield wasreduced when compared with the weed-free check ifweeds were allowed to reach 23 cm in height or more(V5 growth stage), whereas in the untreated check 96%yield loss occurred (Figure 1).
Yield losses, in 2001, also occurred when glyphosatewas applied to weeds 5 cm in height (V3 growth stage)(Figure 1). Losses at the 5-cm timing may be attributedto interference from weeds emerging after glyphosateapplication. This was the only time during this 4-yr studythat weeds emerging after glyphosate applications couldbe implicated in reducing corn yield. This may be duein part to the drought conditions that occurred during thesummer months of 2001. Under these conditions, com-petition for soil moisture by late-emerging weeds mayhave reduced corn yield when glyphosate was applied toweeds 5 cm in height (V2 and V3 corn at Clarksvilleand East Lansing, respectively). In the first 3 yr of thisstudy, weeds that emerged after glyphosate applicationsdid not negatively affect corn yield possibly because soilmoisture was more plentiful due to higher rainfall duringthese years.
Sequential glyphosate applications in corn had no ef-fect on yield. Corn yield after one glyphosate applicationwas always similar (at P , 0.10) to corn yield aftersequential glyphosate applications even in 2001, whenlate-emerging weeds reduced corn yield (data notshown). This is consistent with the findings of Murphyet al. (1996), who found that late-emerging weeds wereless likely to reduce corn yield than weeds emergingwith corn.
At Clarksville in 1998 and 1999, corn yield, combinedover row spacings, was not reduced by weed interferencewhen weeds were allowed to reach 15 cm in height (V6to V7 growth stage) (Figure 2). Yield losses occurredonly in the untreated check, and there was no significantyield increase when corn was planted in narrow rows(data not shown).
Delaying POST herbicide applications in corn increas-es the risk of yield loss due to weed competition (Careyand Kells 1995; Knake and Slife 1969; Tharp and Kells1999). Results from this study showed that as glyphosateapplication was delayed, yield losses occurred in all 4yr of this study. Planting corn in narrow rows increasedyield in three of 4 yr. This research supports the findingsof others that planting in narrow row spacings does in-crease yield (Lutz et al. 1971; Murphy et al. 1996; Niel-sen 1988; Porter et al. 1997; Stickler 1964; Widdicombeand Thelen 2002). However, when corn was grown un-der high levels of weed interference (1999), the yieldadvantage diminished as glyphosate applications weredelayed.
Soybean Yield. In 1998 and 1999 at East Lansing, whensoybean was grown under high levels of weed interfer-ence (low rainfall and high weed density), there was asignificant interaction between time of glyphosate appli-cation and row spacing for soybean yield. As in corn,yield advantages existed for soybean planted in narrowrows, but they were more susceptible to early-seasonweed interference than soybean planted in wide rowspacings (Figure 3). In 1998, weed interference resultedin yield losses when weeds were allowed to reach 23 cmor taller (V5 growth stage) with soybean planted in 19-cm rows. Yield losses occurred only in the untreatedchecks when soybean was planted in 38- or 76-cm rows(.V5 growth stage). In 1999, weed interference reducedsoybean yield from the weed-free check when glyphos-
DALLEY ET AL.: EFFECT OF GLYPHOSATE APPLICATION TIMING AND ROW SPACING
172 Volume 18, Issue 1 (January–March) 2004
Figure 3. Effect of glyphosate application timing on soybean yield at EastLansing. In 1998, glyphosate was not applied to 5- and 30-cm-tall weeds insoybean. WF, weed free; UNT, untreated. Data from 2000 and 2001 wereaveraged across row spacings. Yield means with different lower-case lettersare significantly different (P , 0.05) comparing treatment timings within rowspacing. Yield means with different upper-case letters are significantly differ-ent (P , 0.10) comparing row spacing within each treatment timing.
Figure 4. Effect of glyphosate application timing on soybean yield at Clarks-ville. WF, weed free; UNT, untreated. Yield means with different lower-caseletters are significantly different (P , 0.05) comparing treatment timings with-in row spacing. Yield means with different upper-case letters are significantlydifferent (P , 0.10) comparing row spacing within each treatment timing.
ate applications were delayed until weeds reached 15 cmor more (V2 growth stage), when soybean was plantedin 19- or 38-cm rows, whereas yield reductions occurredonly in the untreated check when soybean was plantedin 76-cm rows (.V5 growth stage). Yield also was re-duced compared with the weed-free check with soybeanplanted in 19- and 38-cm rows when glyphosate wasapplied to 5-cm-tall weeds. It is unclear why this yieldloss occurred because few or no new weeds emergedafter glyphosate application at this timing (personal ob-servation). In both 1998 and 1999 there was a yield ad-vantage for narrow-row soybean. Soybean planted inboth 19- and 38-cm rows yielded more than soybean in76-cm rows only in the weed-free check and at the gly-phosate application to 10-cm-tall weeds. Soybean plant-ed in 19- and 38-cm rows yielded similarly in all treat-ment timings except the untreated check.
In 2000, when soybean was grown under less-stressfulconditions, yield was reduced by weed interference only
in the untreated check (.V5 growth stage) in all rowspacings, and there was no interaction between applica-tion timing and row spacings (Figure 3). However, soy-bean yields differed significantly across row spacings.When averaged across all treatment timings, soybeanplanted in 19-cm rows yielded 5% more than soybeanplanted in 38-cm rows and yielded 9% more than soy-bean planted in 76-cm rows (data not shown). Soybeanplanted in 38-cm rows yielded 4% more than soybeanplanted in 76-cm rows (data not shown).
The drought conditions of 2001 affected soybean andcorn yields in a similar manner. Under drought condi-tions there was no yield advantage for soybean plantedin narrow rows. Because there was no significant differ-ence between row spacings and no interaction betweenrow spacing and time of glyphosate application, yieldswere averaged across row spacings (Figure 3). Soybeanyield losses due to weed interference occurred only inthe untreated check (.V4 growth stage), where yieldlosses of 54% occurred.
At Clarksville, in 1998, there was a significant inter-action between time of glyphosate application and rowspacing for soybean yield. In the weed-free check, soy-bean planted in 19- and 38-cm rows yielded more thansoybean planted in 76-cm rows (Figure 4). There wereno significant differences in yield due to row spacing atany other treatment timing. When soybean was plantedin 19-cm rows, yield was reduced when weeds were al-
WEED TECHNOLOGY
Volume 18, Issue 1 (January–March) 2004 173
Figure 5. Effect of glyphosate application timing on corn and soybean yieldsat East Lansing. UNT, untreated. Yield means with different lower-case lettersare significantly different (P , 0.10) comparing treatment timings withincrops. Yield means with different upper-case letters are significantly different(P , 0.10) comparing corn and soybean within treatment timings.
lowed to reach 15 cm. When soybean was planted in 38-and 76-cm rows, yield losses occurred only in the un-treated check. In 1999, yield was reduced only in theuntreated check, and there were no differences in yielddue to row spacing (Figure 4). When averaged acrossyears, weed-free soybean planted in 19-cm rows yielded7 and 10% more than soybean planted in 38- and 76-cmrows, respectively, and soybean planted in 38-cm rowsyielded 2% more than soybean in 76-cm rows (data notshown).
Delaying glyphosate application in narrow-row soy-bean resulted in yield losses in the first 2 yr of this studyat both locations but not during the final 2 yr of thisresearch. Differences in the results between these yearswere likely related to differences in the degree of weedinterference related to the environmental conditions andweed density in which the soybean was grown. Van Ges-sel et al. (2000) also reported that under severe weedinterference, yield reductions occurred that did not occurunder lower levels of weed interference. Krausz et al.(2001) reported that delaying glyphosate applications insoybean until weeds reach 30 cm in height did not re-duce yield when soil moisture was adequate. Our resultsshowed that soybean yield was first reduced by early-season weed competition at smaller weed heights whensoybean was planted in narrow rows (19 and 38 cm) thanin wide rows (76 cm). Soybean planted in 19- and 38-cm rows yielded similarly in most cases; however, soy-bean planted in 19-cm rows yielded more than soybeanplanted in 76-cm rows in three of 4 yr in the weed-freecheck.
Corn and Soybean Yields. Yield results from 1998were omitted because of yield losses in the weed-freecheck likely associated with light measurements taken inthese plots. In 1999, weed interference reduced cornyield more than soybean yield when weeds were allowedto reach 23 cm or more (Figure 5). In 2000, there wereno differences in yield loss comparing corn and soybeanat any of the glyphosate application timings, but soybeanyield was reduced more than corn yield in the untreatedcheck. In 2001, corn yield was reduced more by weedinterference than was soybean yield when glyphosatewas applied to weeds that were 5 or 23 cm in height.Corn yield also was reduced more than soybean yield inthe untreated check (Figure 5). At Clarksville, there wereno significant differences between corn and soybeanyields at any treatment timing in 1998 or 1999 (data notshown).
Delaying herbicide applications until weeds reached23 cm or more reduced corn yield to a greater extent
than soybean yield under the highly competitive growingconditions (high weed density and low early-season rain-fall) that occurred in 1999 (Figure 5). In 2000, whengrowing conditions were less competitive, corn and soy-bean yields responded similarly across all glyphosatetreatment timings. In 2001, corn yield was reduced morethan soybean yield by early-season weed interference atthe 23-cm weed height application timing and by late-season weed interference when glyphosate was appliedat the 5-cm weed height. In 2000, when soybean yieldwas reduced more than corn yield in the untreated check,corn yield was reduced only 22% compared with theweed-free check, whereas soybean yield was reduced39% compared with the weed-free check. In 1999 and2001, corn yield was reduced 83 and 92%, respectively,and soybean yield was reduced 63 and 58%, respective-ly, compared with the weed-free check. These findingssuggest that under highly competitive growing condi-tions corn is less tolerant of weed interference than soy-bean. Knake and Slife (1969) also found that early-sea-son weed competition had greater effects on corn thanon soybean. They found that the greatest competitiveeffect of weeds on soybean appears to occur after thestart of the reproductive stage and that early-season weedcompetition delayed corn tassel emergence.
Our results indicated that under highly competitive
DALLEY ET AL.: EFFECT OF GLYPHOSATE APPLICATION TIMING AND ROW SPACING
174 Volume 18, Issue 1 (January–March) 2004
Table 5. Adjusted R2 values for corn and soybean yields and application timing parameters.
Weed densityDAE by CLV-stage by CLGDDAE (base 42) by CLCrop height by CLWeed height by CL
0.22780.07340.09490.10910.11950.1854
0.240620.0023
0.00470.00980.01910.0725
0.044520.013020.012620.012320.011720.0061
0.185520.0076
0.00930.00460.02340.0417
0.19900.15670.19660.18260.18220.2354
DAE by weed densityV-stage by weed densityGDDAE (base 42) by weed densityCrop height by weed densityWeed height by weed density
0.26820.27310.28200.29180.3012
0.25660.25980.28170.29890.3437
0.02640.02790.03800.03230.0352
0.17670.31470.28780.32200.3829
0.36790.38340.38780.40290.4430
a Abbreviations: CL, Michigan competitive load; GDDAE, growing degree days after crop emergence; DAE, days after crop emergence.
growing conditions (high weed density and less than nor-mal rainfall) weeds needed to be controlled as early as5 cm in height (15 to 23 DAE, at the V3 corn leaf stage)to prevent yield loss in narrow-row corn and beforeweeds exceeded 15 cm (21 to 30 DAE, at the V2 to V3soybean leaf stage) to prevent yield loss in narrow-rowsoybean. When corn was planted in wide rows (76 cm),weeds needed to be controlled before they reached 10cm in height (19 to 26 DAE, at the V4 corn leaf stage).Weeds reaching 30 cm in height (31 to 38 DAE, at theV4 to V5 soybean leaf stage) in wide-row soybean didnot reduce soybean yield during any year in which thisstudy was conducted.
Predicting Yield Loss. Herbicide application timingsare frequently based on a various parameters such asdays or weeks after crop emergence (Horn and Burnside1985), crop height or growth stage (Hall et al. 1992; VanGessel et al. 2000), or weed height (Knake and Slife1969; Krausz et al. 2001). In this study, timing of gly-phosate application was based on weed height. However,crop growth stage, crop height, DAE, and GDDAE alsowere recorded (Tables 1 and 2). Competitive load11 alsowas calculated using weed densities and competitive in-dices for weeds in corn and soybean (Table 3). Simplelinear regression analysis of crop yield response to theseapplication timing parameters resulted in fairly low co-efficients of determination (R2). Of the five applicationtiming parameters (weed height, crop height, cropgrowth stage, DAE, and GDDAE), the highest R2 values
11 Weedsoft, 362 Plant Sciences Building, University of Nebraska, Lincoln,NE 68583-0915.
were obtained using weed height as the independent var-iable (Table 5).
The weed height at which yield losses first occurredvaried considerably during the different growing sea-sons. Other factors, such as weather, crop growth stage,weed density, or weed composition, may have signifi-cantly influenced yield results. Yield losses have beenshown to increase with increased weed density. Whenthe product of weed density and application timing pa-rameters was used, R2 values improved considerably (Ta-ble 5). The best fit was obtained by using the product ofweed height (cm) and weed density (weeds/m2) as theindependent variable for regression analysis of corn andsoybean yields (Figure 6). These regressions had R2 val-ues of 0.31, 0.35, and 0.05 for soybean planted in 19-,38-, and 76-cm rows and 0.39 and 0.45 for corn plantedin 38- and 76-cm rows, respectively.
The R2 values were not improved by using competi-tive load values in place of weed density (Table 5). Intheory, competitive load provides the ability to sum thedensity and competitiveness of the individual weed spe-cies that accounts for the differences in competitive abil-ity of different weed species. The competitive indicesused in calculating the competitive load are derived fromresults of studies comparing the effects of season-longweed interference and may not be appropriate for cal-culating yield losses due to weed competition early inthe growing season. Competitive indices must continueto be refined to more accurately predict the effects ofweed interference.
Weed density appears to be one of the most importantfactors in determining when yield losses begin to occur
WEED TECHNOLOGY
Volume 18, Issue 1 (January–March) 2004 175
Figure 6. Linear regression of yield and timing of glyphosate applicationbased on the product of weed density and weed height of corn planted in 38-and 76-cm rows and soybean planted in 19-, 38-, and 76-cm rows at EastLansing (1998, 1999, 2000, and 2001) and Clarksville (1998, 1999, and 2001).
due to weed interference. Combining weed density withweed height, crop height, crop growth stage, DAE, orGDDAE resulted in a better linear fit than when any ofthese parameters were analyzed alone using linear re-gression. In this study, the combination of weed heightand weed density was the best predictor of yield loss forboth corn and soybean. This finding suggests that usinga single variable such as weed height or crop growthstage as a basis for timing glyphosate applications with-out considering weed density is a flawed approach. Highweed densities increase the risk of yield loss and mustbe considered when determining the appropriate timingfor total POST herbicide applications such as glyphosate.Further research needs to be conducted on the effects ofweed density on early-season weed competition.
Corn and soybean planted in narrow rows were af-fected differently from corn and soybean planted in widerows. As row spacing narrowed in both crops, the influ-ence of early-season weed interference increased. Thisshowed that corn and soybean planted in narrow rows
were more susceptible to yield loss from weed interfer-ence than corn and soybean planted in wide rows. TheR2 values for soybean planted in 76-cm rows were verylow in all regression analyses because significant yieldlosses did not occur at any of the treatment timings, re-sulting in a slope approaching zero. However, yield loss-es did begin to occur at some point after the final gly-phosate application to 30-cm weeds (V5 growth stage)because yields in the untreated checks were reduced con-siderably.
Proper timing of glyphosate applications in glyphos-ate-resistant crops is necessary to avoid yield losses dueto early-season weed interference. The results of this re-search showed that corn and soybean were more suscep-tible to early-season weed interference when grown un-der stressful growing conditions, such as high weed den-sities or low rainfall. Weed densities varied considerablyin the 4 yr this research was conducted. When high weeddensities occurred, yield losses were observed at earlierapplication timings (smaller weed heights) than whenlow weed densities occurred. The combination of weedheight and weed density appeared to be the best predictorof yield loss due to weed interference for both corn andsoybean.
The results also showed that corn and soybean plantedin narrow rows growing with high levels of weed inter-ference were more susceptible to yield losses comparedwith wide-row corn and soybean. However, when weedswere controlled early season, corn and soybean plantedin narrow rows yielded more than those in wide rows inthree of 4 yr.
LITERATURE CITED
Ateh, C. M. and R. G. Harvey. 1999. Annual weed control by glyphosate inglyphosate-resistant soybean (Glycine max). Weed Technol. 13:394–398.
Board, J. E. and B. G. Harville. 1993. Soybean yield component responsesto a light interception gradient during the reproductive period. Crop Sci.33:772–777.
Board, J. E. and B. G. Harville. 1996. Growth dynamics during the vegetativeperiod affects yield of narrow-row late-planted soybean. Agron. J. 88:567–572.
Bullock, D., S. Khan, and A. Rayburn. 1998. Soybean yield response to nar-row rows is largely due to enhanced early growth. Crop Sci. 38:1011–1016.
Bullock, D. G., R. L. Nielsen, and W. E. Nyquist. 1988. A growth analysiscomparison of corn grown in conventional and equidistant plant spacing.Crop Sci. 25:254–258.
Carey, J. B. and J. J. Kells. 1995. Timing of total postemergence herbicideapplications to maximize weed control and corn yield. Weed Technol. 9:356–361.
Classen, M. M. and R. H. Shaw. 1970. Water deficit effects on corn. II. Graincomponents. Agron. J. 62:652.
Egli, D. B. 1994. Mechanisms responsible for soybean yield response to equi-distant patterns. Agron. J. 86:1046–1049.
Ethredge, W. J., Jr., D. A. Ashley, and J. M. Woodruff. 1989. Row spacingand plant population effects on yield components of soybean. Agron. J.81:947–951.
DALLEY ET AL.: EFFECT OF GLYPHOSATE APPLICATION TIMING AND ROW SPACING
176 Volume 18, Issue 1 (January–March) 2004
Hall, M. R., C. J. Swanton, and G. W. Anderson. 1992. The critical period ofweed control in grain corn (Zea mays). Weed Sci. 40:441–447.
Horn, P. W. and O. C. Burnside. 1985. Soybean growth as influenced byplanting date, cultivation, and weed removal. Agron. J. 77:793–795.
Ikeda, T. 1992. Soybean planting patterns in relation to yield and yield com-ponents. Agron. J. 84:923–926.
Knake, E. L. and F. W. Slife. 1969. Effect of time of giant foxtail removalfrom corn and soybeans. Weed Sci. 17:281–283.
Krausz, R. F., G. Kapusta, and J. L. Matthews. 1996. Control of annual weedswith glyphosate. Weed Technol. 10:957–962.
Krausz, R. F., B. G. Young, G. Kapusta, and J. L. Matthews. 2001. Influenceof weed competition and herbicides on glyphosate-resistant soybean(Glycine max). Weed Technol. 15:530–534.
Lutz, J. A., H. M. Camper, and G. D. Jones. 1971. Row spacing and popu-lation effects on corn yield. Agron. J. 63:12–14.
Murphy, S. D., Y. Yakubu, S. F. Weise, and C. J. Swanton. 1996. Effect ofplanting pattern and inter-row cultivation on competition between corn(Zea mays) and late emerging weeds. Weed Sci. 44:856–870.
Nelson, K. A. and K. A. Renner. 1999. Weed management in wide- andnarrow-row glyphosate resistant soybean. J. Prod. Agric. 12:460–465.
Nielsen, R. L. 1988. Influence of hybrids and plant density on grain yield andstalk breakage in corn grown in 15-inch row spacing. J. Prod. Agric. 1:190–195.
Padgette, S. R., K. H. Kolacz, X. Delannay, et al. 1995. Development, iden-tification, and characterization of a glyphosate-tolerant soybean line.Crop Sci. 35:1451–1461.
Paszkiewicz, S. R. 1997. Narrow row width influence on corn yield. In Pro-
ceedings 51st Annual Corn and Sorghum Research Conference; Chicago,IL. American Seed Trade Association, Washington, DC. Pp. 130–138.
Porter, P. M., D. R. Hicks, W. E. Lueschen, J. H. Ford, D. D. Warnes, and T.R. Hoverstad. 1997. Corn response to row width and plant population inthe northern corn belt. J. Prod. Agric. 10:293–300.
Stickler, F. C. 1964. Row width and plant population studies with corn. Agron.J. 56:438–441.
Tharp, B. E. and J. J. Kells. 1999. Influence of herbicide application rate,timing, and interrow cultivation on weed control and corn (Zea mays)yield in glufosinate-resistant and glyphosate-resistant corn. Weed Tech-nol. 13:807–813.
Tharp, B. E., O. Schabenberger, and J. J. Kells. 1999. Response of weeds toglufosinate and glyphosate. Weed Technol. 13:542–547.
Van Gessel, M. J., A. A. Ayeni, and B. A. Majek. 2000. Optimum glyphosatetiming with and without residual herbicides in glyphosate resistant soy-bean (Glycine max) under full-season conventional tillage. Weed Tech-nol. 14:140–149.
Weaver, S. E., M. J. Kropff, and R.M.W. Groeneveld. 1992. Use of ecophys-iological models for crop-weed interference: the critical period of weedinterference. Weed Sci. 40:302–307.
Westgate, M. E., F. Forcella, D. C. Reicosky, and J. Somsen. 1997. Rapidcanopy closure for maize production in the northern US corn belt: ra-diation-use efficiency and grain yield. Field Crops Res. 49:249–258.
Widdicombe, W. D. and K. D. Thelen. 2002. Row width and plant densityeffects on corn grain production in the Northern corn belt. Agron. J. 94:1020–1023.
Wiggins, R. G. 1939. The influence of space and arrangement on the produc-tion of soybean plants. J. Am. Soc. Agron. 31:314–321.
