ArticlePDF Available

Two methods of composting gin trash


Abstract and Figures

The ginning of cotton (Gossypium hirsutum L.) results in the accumulation of approximately 90 kg of gin trash per bale of cotton ginned. In the past, disposal of raw gin trash was by burning, land application and feeding to livestock, but problems with clean air standards, weed seeds and diseases, and chemical residues, respectively, make each of these methods unacceptable. Composting the gin trash would alleviate certain problems associated with land application on farm fields. Experiments were conducted to investigate windrow composting. A split plot experiment with five reps was initiated to evaluate turning times, nitrogen (N) fertilization and bacterial inoculation. Composite samples were collected and analyzed for nutrients and selected chemicals. Another experiment was established to compare timing and physical methods of turning windrows. The Lipsey®-gin-trash-composting system was investigated by sampling three compost piles at three different cotton gins. The chemical composition and weed seed germination were investigated. Results indicate that windrow composting does not solve the weed seed or plant disease problem; otherwise, the product was satisfactory. The Lipsey®-gin-trash-composting system resulted in a pile of material whose outside 0 to 15 cm depths contained viable weed seeds and disease organisms, but none survived below this exterior. This system resulted in an incomplete composting material with offensive odors.
Content may be subject to copyright.
Division of Agriculture University of Arkansas
July 1998 Special Report 186
This proceedings and the companion Southern Conservation Tillage Conference for Sustainable Agriculture are activi-
ties of the Southern Extension and Research Activity - Information Exchange Group 20 (SERA-IEG-20), which is
sponsored by the University of Arkansas, the Southern Association of Agricultural Experiment Station Directors, the
Southern Association of Agricultural Extension Service Directors and the Cooperative State Research, Education, and
Extension Service (CSREES).
Agricultural Experiment Station, University of Arkansas Division of Agriculture, Fayetteville. Milo J. Shult, Vice
President for Agriculture and Director; Charles J. Scifres, Associate Vice President for Agriculture. PS1.5M798PM
The Arkansas Agricultural Experiment Station follows a nondiscriminatory policy in programs and employment.
Proceedings of the
21st Annual Southern
Conservation Tillage Conference
for Sustainable Agriculture
North Little Rock, Arkansas
July 15-17, 1998
Terry C. Keisling, Editor
Arkansas Agricultural Experiment Station
Fayetteville, Arkansas
Financial Supporters of the 21st Annual
Southern Conservation Tillage Conference
for Sustainable Agriculture
American Cyanamid
DuPont Agricultural Products
Natural Resources Conservation Service
Zeneca Agricultural Products
University of Arkansas
Cooperative Extension Service
Agricultural Experiment Station
1890 Cooperative Extension Program
Conservation tillage, especially no-till, gained greater acceptance during the decades of the 1960s and 1970s. This
acceptance coincided with the availability of herbicides that could substitute for mechanical cultivation for weed
control. Highly erodible locations were usually the first to implement conservation practices.
Conservation tillage generally reduces erosion, conserves energy costs associated with tillage operations and modi-
fies soil-water relationships. Conservation tillage often requires greater herbicide use to obtain acceptable weed
control. Under reduced tillage scenarios, applied lime and fertilizer tend to concentrate in the surface few inches of
soil. Greater capture of rainfall and fast transmission of water via large pores to greater depths may pose an increased
potential for ground water contamination with pesticides and nitrates. In some cases, continual cropping without
mechanical tillage has resulted in increased surface soil compaction.
Conservation tillage issues that evolved during the 1980s included effective herbicide and fertilizer use, proper soil
sampling techniques, insect and disease management, crop residue management, soil-water relations, surface and
ground water protection and profitability of crop production. Numerous production problems have been addressed, and
various solutions are being tested. As conservation technology improves, its acceptance continues to increase.
During the 1990s, as much as 35% of the crop land in the United States is being farmed with some kind of
conservation tillage practice. The advent of bioengineering of herbicide-resistant crops has made weed control in
conservation tillage easier. With adaptation of conservation tillage, equipment that addresses various problems that
occur when using conservation tillage has been developed in farm shops and then been offered commercially by
equipment companies.
The 1998 conference theme, “MEETING THE CHALLENGES” was chosen for its focus on removing the barriers
of further adaptation of conservation tillage while sustaining that which is in place. To be sustainable requires that a
balance among profitable agriculture production, socially acceptable practices and environmentally sound practices be
achieved. The 1998 conservation tillage conference continues to provide a communication link among various agencies
and personnel interested in improved natural resource management. We here at the University of Arkansas appreciate
the opportunity to host this annual conference and to facilitate the adaptation of conservation tillage technology.
Stan L Chapman
Extension Soil Specialist and Agronomy Section Leader
Cooperative Extension Service
University of Arkansas
P.O. Box 391
Little Rock, Arkansas 72203
Terry C. Keisling
Professor of Agronomy
University of Arkansas Northeast Res. & Ext. Ctr.
P.O. Box 48
Keiser, Arkansas 72361
Philip J. Bauer, John A. DuRant and James R.
Frederick ...................................................................9
W.J. Busscher and P.J. Bauer ................................. 13
T.A. Castillo, T.K. Keisling and L.R. Oliver .......... 16
M.B. Daniels, S.L. Chapman, R. Matlock
and A. Winfrey ....................................................... 19
Ernest H. Flint, Jr., Glover B. Triplett, Jr., Seth M.
Dabney, William H. Batson, Dawn S. Luthe and
Clarence E. Watson................................................ 22
R.N. Gallaher and R. McSorley ............................. 25
R.N. Gallaher, J.D. Greenwood and
R. McSorley ........................................................... 28
E.C. Gordon, T.C. Keisling, D.M. Wallace,
L.R. Oliver and C.R. Dillon ................................... 32
E.C. Gordon, T.C. Keisling, L.R. Oliver
and Carl Harris ....................................................... 38
E.M. Holman, A.B. Coco and R.L. Hutchinson ..... 42
D.D. Howard, M.D. Mullen and M.E. Essington ... 46
Michael D. Hubbs, D.W. Reeves and Charles C.
Mitchell Jr. ............................................................. 50
J.R. Johnson and J.R. Saunders .............................. 55
T.C. Keisling, E.C. Gordon, G.M. Palmer
and A.D. Cox .......................................................... 58
R.A. Klerk, J.D. Beaty, L.O. Ashlock,
C.D. Brown and T.E. Windham ............................... 59
H.J. Mascagni, Jr. and D.R. Burns.......................... 62
Marilyn R. McClelland, M. Cade Smith
and Preston C. Carter ............................................. 65
J.M. McKimmey and H.D. Scott ........................... 70
Ronald Morse......................................................... 79
L.R. Oliver and M.T. Barapour ............................... 8 3
Robert G. Palmer ................................................... 87
J.R. Saunders and J.R. Johnson .............................. 90
P.J. Wiatrak, D.L. Wright, J.A. Pudelko, B. Kidd
and W. Koziara ....................................................... 92
D.L. Wright, P.J. Wiatrak, D. Herzog, J.A. Pudelko
and B. Kidd ............................................................. 95
Philip J. Bauer, John A. DuRant and James R. Frederick1
There is considerable variability for lint yield within
cotton (Gossypium hirsutum L.) fields in the south-
eastern Coastal Plain. The objective of this experi-
ment was to determine if soil management techniques and
in-furrow application of an insecticide/nematicide influ-
ence the amount of variability in cotton yield and fiber
properties. Treatments in the study were tillage (conser-
vation vs. conventional) and aldicarb application (1.07 lb
ai/acre vs. none). In 1997, ‘DPL Acala 90’ was planted
into large plots (ranging in length from approximately 400
to 800 ft, plots were six 38-in.-wide rows) that spanned
across several soil map units. Two harvesting methods
were used to determine variability. First, the large plots
were subdivided into 44-ft-long sections, and two of the
rows in each section were harvested with a spindle picker.
Second, a 6-ft sample was hand-harvested from each of
three soil map units (Bonneau sand, Eunola loamy sand
and Norfolk loamy sand) within each plot. Neither aldicarb
application nor tillage system affected the variability for
yield or micronaire among the machine-harvested samples.
Variability for fiber length was less in conservation tillage
than in conventional tillage only when aldicarb was ap-
plied. For fiber strength, conservation tillage had lower
variability than conventional tillage for the plots without
aldicarb. Soil map unit was responsible for much of the
variation in yield, with the Bonneau sand having lower
yield than the other two soil map units. Variability for
fiber properties was less than variability for yield.
A large amount of variation in cotton growth and pro-
ductivity can occur within the cotton fields of the south-
eastern Coastal Plain. One of the largest sources appears
to be variation due to soil map unit. Fields in this region
generally have many soil map units and a range of physical
and chemical properties that influence crop growth (Karlen
et al., 1990). The primary productivity differences among
soil map units may be in differences in ability to supply
water to crops. Sadler et al. (1998) found a significant
relationship between canopy minus air temperature and
soil map unit in corn (Zea mays) during a severe water
1P.J. Bauer,USDA-ARS, Coastal Plains Soil, Water, and Plant Research
Center, Florence, South Carolina. J.A. Durant and J.R. Frederick, Clemson
University, Florence, South Carolina.
deficit year, which implied that soil physical differences
caused differences in water stress.
Many of the benefits of conservation tillage, especially
when used with adequate residue cover, are related to im-
proving soil water conditions. Benefits often cited include
increased rainfall infiltration, reduced runoff and reduced
evaporation from the soil surface. Thus, conservation till-
age techniques may reduce the amount of field variability
for cotton yield by reducing the amount of in-field vari-
ability for soil water.
Besides soil map unit, pest infestations are a source of
variability in the southeastern Coastal Plain cotton fields.
Although seldom random, infestations of weeds, insects
and nematodes do not tend to be uniformly distributed
throughout a field. Though pests are rarely uniformly dis-
tributed, pest control measures are usually applied uni-
formly throughout a field. Part of the reason for this is
the uncertainty of where pest infestations will occur. Also,
there is very little spatial data available on the efficacy of
pest control products.
A six-year study was established in the fall of 1996
with the overall objective to determine the effects of resi-
due amount, tillage system and in-furrow insecticide ap-
plication on cotton yield and fiber properties. In this re-
port, we describe our results from the first year of con-
verting a field to a conservation tillage production system.
The objective is to determine if soil management tech-
niques and in-furrow application of an insecticide/nemati-
cide influence the amount of variability in cotton yield
and fiber properties.
Seven acres of a 40-acre field at Clemson University’s
Pee Dee Research and Education Center near Florence,
South Carolina, were used for the experiment. The area
was chosen because of the diversity in soil map units and
the ability to have at least two soil map units represented
in each plot. Treatments were tillage (conventional or con-
servation) and in-furrow insecticide/nematicide applica-
tion (aldicarb or none). Experimental design was split-
plot with main plots in a randomized complete block. There
were three blocks. Main plots were the tillage treatments,
and subplots were the in-furrow insecticide application
treatments. Main plot size was twelve 38-in.-wide rows
that ranged in length from approximately 400 ft to more
than 800 ft. Six of the rows received an in-furrow applica-
tion of 1.07 lb ai/acre aldicarb, while the other six were
planted without insecticide/nematicide protection to serve
as controls.
In previous years, the field was in a two-year rotation
of corn followed by winter wheat (Triticum aestivum)
double-cropped with soybean (Glycine max). Corn was
grown in the field during the summer of 1996. Following
corn harvest, stalks were mowed. The experiment was
originally designed to include a rye cover crop (both with
and without tillage) treatment. Rye was planted 20 No-
vember 1996, but because of poor cover crop growth,
these plots were pooled with the no-cover-crop main plots
for this analysis. In the spring of 1997, paraquat was ap-
plied to the conservation tillage plots while the conven-
tional tillage plots were disked and then smoothed with a
harrow equipped with S-shaped tines and rolling baskets.
On 2 May, a paratill with shanks spaced 26 in. apart was
used to deep-till the entire experimental area to a depth of
16 in.
Cotton (‘Deltapine Acala 90’) was planted 7 May using
a four-row planter equipped with waved coulters. Seeding
rate was four seeds per row-ft. Preemergence herbicides
(fluometuron and pendimethalin) were applied 8 May.
Post-emergence herbicides included pyrithiobac, cyanazine
and monosodium methanearsonate. All herbicides were
applied at recommended rates. Plant nutrients (other than
N) were broadcast applied before cotton planting at rates
based on soil test results and Clemson University Coop-
erative Extension Service recommendations. All N was
side-dress applied in a split application, with 40 lb N/acre
being applied 13 May and 40 lb N/acre applied 20 June.
All N applied was NH4NO3.
Two methods of harvest were used to assess the yield
and fiber property variability. The first method involved
separating each subplot into 50-ft-long sections and re-
moving plants from 3 ft of row from each end of the
sections so that the harvested area within each section was
44 ft long. A two-row spindle picker was used to harvest
two of the rows in each section. A grab sample of
seedcotton from each harvest bag was collected at harvest
for fiber property determinations. The second method in-
volved hand-harvesting 6 ft of row from individual soil
map units within each plot. The map units chosen were
Bonneau sand (BoB; loamy, siliceous, thermic Arenic
Paleudult), Eunola loamy sand (EuB; Fine-loamy, siliceous,
thermic Aquic Hapludult) and Norfolk loamy sand (NoA;
fine-loamy, siliceous, thermic Typic Kandiudult). All three
soil map units were present in all plots in two of the
blocks. In the other block, the EuB soil map unit was in
each main plot, while the BnA soil map unit was present in
only one of the four main plots, and the NoA map unit was
present in only three of the four main plots. All seedcotton
samples were ginned on a 10-saw laboratory gin. Samples
of the lint samples were sent to Star-Lab, Inc (Knoxville,
Tennessee) for HVI fiber property determinations.
Bartlett’s F test for homogeneity of variance was con-
ducted to determine if the amount of variability differed
between conventional tillage and conservation tillage for
both levels of aldicarb application. Since the experimental
design was split-plot with main plots in a randomized com-
plete block design, variance components for each subplot
treatment consisted of variation due to blocks and to
within-plot variation. Therefore, an analysis of variance
for treatment combination (tillage x aldicarb) was con-
ducted to remove the variance component due to blocks,
and the residual mean square was used as the estimate of
σ2 for conducting Bartlett’s F test. For the hand-harvested
samples, data were analyzed by analysis of variance using
the general linear models (PROC GLM) procedure of SAS.
Estimates of variance and Bartlett’s F test for
heterogeniety of variance among the machine-picked
samples for lint yield, fiber length, strength and micronaire
are given in Table 1. The amount of variability for cotton
yield did not differ between conventional tillage and con-
servation tillage either with or without aldicarb applica-
tion (Table 1). Similarly, variability did not differ for
micronaire between the tillage systems either with or with-
out aldicarb. Heterogeneity of variance was found for both
fiber length and fiber strength. In both cases, the conser-
vation tillage had lower variance than did conventional till-
age. For fiber length, variance was lower for conservation
tillage than for conventional tillage when aldicarb was ap-
plied (Table 1). For fiber strength, variance of the conser-
vation tillage was less when aldicarb was not applied.
For the machine-harvest sampling method, a significant
(P < 0.10) tillage x aldicarb interaction occurred for lint
yield (Table 2). With aldicarb, the conventional and con-
servation tillage production systems had similar yield
(Table 2), averaging 859 lb lint/acre. The interaction was
caused by magnitude differences between aldicarb-treated
and untreated cotton within each tillage system. In conser-
vation tillage, yields of cotton without aldicarb were only
131 lb/acre less than the cotton treated with aldicarb. In
conventional tillage, the difference between aldicarb-
treated and untreated was 212 lb lint/acre (Table 2). Early-
season counts indicated that thrips populations were less
in the conservation tillage than in the conventional (data
not shown). Only small, and probably inconsequential, mean
differences among treatments occurred for fiber proper-
ties with the machine harvest sampling method. As ex-
pected, it appears that much of the within-plot variability
found with the machine-harvest method was due to soil
map unit.
Yield and fiber properties from the hand-harvested
samples are given in Table 3. Averaged over tillage sys-
tems and aldicarb levels, lint yields were 694 lb/acre for
the Bonneau, 913 lb/acre for the Eunola and 1020 lb/acre
for the Norfolk. The average yield increase due to aldicarb
was 158 lb lint/acre. The micronaire response was similar
to yield, with lower micronaire occurring on the Bonneau
soil map unit than on the other two and aldicarb-treated
cotton having higher micronaire than untreated. As for the
machine-harvested samples, variability for fiber length and
strength was small, even when treatment means were sig-
nificantly different. Notably, the cotton produced with con-
servation tillage on the Bonneau soil grown without
aldicarb was substantially lower for yield and fiber quality
than the other treatment combinations in the experiment.
Although the tillage x aldicarb x soil map unit interac-
tion was not significant for lint yield (P = 0.198), inspec-
tion of the means provides some indication of why the
tillage x aldicarb interaction occurred for yield with the
machine-picked data. As discussed earlier, yield reduc-
tions without aldicarb were less in the conservation tillage
production system than in the conventional tillage system.
Aldicarb did not increase yield for the Eunola and Norfolk
soils in the conservation tillage system but resulted in a
substantial yield increase on these two soils in conven-
tional tillage (Table 3). For the Bonneau soil, aldicarb
treatment increased yield in both the conservation and con-
ventional tillage treatments. Unfortunately, insect pest
monitoring was not conducted on an individual soil map
unit basis in 1997.
These preliminary data suggest that there can be sub-
stantial yield and fiber property variation within fields for
cotton in the southeastern Coastal Plain. Additionally, al-
though within-field variation for yield was not reduced
with conservation tillage, conservation tillage did decrease
the within-plot variation for fiber length and strength. Ap-
plication of aldicarb did not reduce within-plot variability,
nor did it have much of an effect on variability among soil
types. More in-depth monitoring of insect and nematode
pests is planned.
We thank Bobby Fisher, Van Atkinson and Gene Taylor
for technical assistance and Ellen Whitesides for helping
prepare the manuscript. Mention of a trademark, propri-
etary product, or vendor does not constitute a guarantee
or warranty of the product by the USDA or Clemson Uni-
versity and does not imply its approval to the exclusion of
other products or vendors that may also be suitable.
Karlen, D.L., E.J. Sadler and W.J. Busscher. 1990. Crop yield
variation associated with Coastal Plain soil map units. Soil Sci.
Soc. Am. J. 54:859-865.
Sadler, E.J., W.J. Busscher, P.J. Bauer and D.L. Karlen. 1998. Spatial
scale requirements for precision farming inferred from
observations in the southeastern USA. Agron. J. (inpress).
Steel, R.G.D., and J.H. Torrie. 1980. Principles and procedures of
statistics. Second Edition. McGraw-Hill Book Co., New York,
New York.
Table 1. Estimates of variance for yield and selected fiber properties of each tillage x aldicarb combination in the experiment
and Bartlett’s F test for homogeneity of variance. Estimates are for the machine-harvested samples.
Estimate of σ2
Tillage Aldicarb n Yield Length Strength Micronaire
Conservation Yes 54 25112 0.00050 0.9222 0.0767
No 54 20665 0.00052 0.9820 0.0629
Conventional Yes 58 21976 0.00086 1.3951 0.0688
No 58 25963 0.00054 1.7355 0.0712
Bartlett’s F-test Values for Homogeneity of Variance between Tillage Systems
Yes 1.14 1.72* 1.51 1.11
No 1.26 1.04 1.77* 1.13
*Indicates F value significant at
= 0.05 (F values for determination of significance were estimated from F table values of F0.05 40,40 = 1.69 and F0.05 60,60
=1.53 [Steel and Torrie, 1980]).
Table 2. Average cotton yield and selected fiber properties as affected by tillage
and aldicarb application. Data are from machine-picked samples.
Tillage Aldicarb Yield Fiber Length Fiber Strength Micronaire
lb lint/acre in. g/tex units
Conservation Yes 849 1.12 30.0 4.1
No 718 1.11 30.3 4.1
Conventional Yes 868 1.12 30.3 4.2
No 656 1.11 30.4 4.1
Significance Level (Prob > F Value) From Analysis of Variance
Tillage 0.788 *** *** 0.469
Aldicarb <0.001 *** *** 0.213
Tillage x Aldicarb 0.066 *** *** 0.859
*** Hypothesis testing for these variables is invalid because of heterogeniety of variance.
Table 3. Average cotton yield and selected fiber properties as affected by tillage, aldicarb application
and soil map unit. Data are from hand-harvested samples.
Tillage Aldicarb Soil Map Unit Yield Fiber Length Fiber Strength Micronaire
lb lint/acre in. g/tex units
Conservation Yes Bonneau 795 1.09 32.6 3.7
Eunola 912 1.11 32.4 4.0
Norfolk 1056 1.12 32.7 4.1
No Bonneau 527 1.07 29.9 3.2
Eunola 908 1.11 32.6 4.2
Norfolk 1030 1.11 32.2 3.8
Conventional Yes Bonneau 785 1.12 33.8 3.7
Eunola 1085 1.12 32.5 4.1
Norfolk 1110 1.13 32.8 4.2
No Bonneau 658 1.10 32.5 3.7
Eunola 749 1.11 32.3 3.9
Norfolk 880 1.09 31.7 3.8
Significance Level (Prob > F Value) From Analysis of Variance
Tillage x Aldicarb
Tillage x Soil
Aldicarb x Soil
Tillage x Aldicarb X Soil
0.704 0.295 0.127 0.259
0.007 0.031 <0.001 0.049
<0.001 0.482 0.775 0.003
0.273 0.622 0.787 0.890
0.736 0.066 0.002 0.460
0.929 0.627 0.012 0.241
0.198 0.357 0.208 0.205
W.J. Busscher and P.J. Bauer1
On sandy coastal subsurface hardpan soils, cover
crops have the potential to prevent erosion and
scavenge nutrients. Our objective was to deter-
mine the effect of cover crops and tillage on soil strength
and cotton yield. Treatments were surface tillage (disked
or none), deep tillage (in-row subsoiled or none) and cover
crop (rye or fallow). Soil strength (cone index) differ-
ences were measured for tillage treatments (deep tilled <
none), depth (higher strength in the pan) and position
across the row (in row < non-wheel track < wheel track).
Lower cone indices were found in the non-tilled rye cover,
suggesting that the cover helped maintain low strengths.
Higher cone indices in the disked treatments suggested
that the disking aided recompaction.
In the southeastern Coastal Plains, winter cover is im-
portant for long-term conservation tillage crop produc-
tion. Cool- and warm-season annual double crops are
needed for successful conservation tillage production of
grain sorghum [Sorghum bicolor (L.) Moench] and soy-
bean [Glycine max (L.) Merrill] on southeastern Pied-
mont sandy loams (Langdale et al., 1990). However, be-
cause of the long southeastern cotton growing season,
double cropping with continuous cotton is not possible
for much of the region. In addition, low organic matter
produced by cotton can leave a field bare for the winter.
Cover crops provide winter cover to improve erosion
control and increase infiltration. They can also scavenge
nutrients and reduce groundwater pollution. Cover crops
might also provide the beneficial rotational effect of double
crops seen by Langdale et al. (1990).
Because of the subsurface root-restricting E horizon
of many Coastal Plain soils, in-row subsoiling is needed
to help roots penetrate into the clay-textured B horizon.
In-row subsoiling provides a narrow, soft zone below the
row that roots can use to penetrate through the E and grow
into the B horizon. By adding organic matter from both
roots and cover, cover crops may also help maintain lower
soil strength.
Our objective was to determine the influence of sur-
face tillage, deep tillage and a rye cover crop on soil
strength and cotton lint yield.
1USDA-ARS, Coastal Plain Soil, Water, and Plant Research Center, Florence,
South Carolina.
In 1990, we established cover crop plots at the Clemson
University Pee Dee Research and Education Center near
Florence, South Carolina. Bauer and Busscher (1993) re-
ported the results from the 1991 and 1992 experiment. In
1993, cotton was grown on the plots but not harvested
because of a drought. All plots were subsoiled in spring
In 1994 and 1995, we changed the treatments to
subsoiling only half the plots. During these two years,
experimental treatments were winter cover (rye and fal-
low), surface tillage (disking and none) and deep tillage
(in-row subsoiling and none). The experimental design was
split-split plot randomized complete block. Main plots
were winter cover, subplots were surface tillage, and sub-
subplots were deep tillage. Subsubplots were 12.7 ft wide
(four 38-in. rows) by 50 ft long. The experiment had four
replicates. The soil was a Norfolk sandy loam (fine, loamy,
siliceous, thermic, Typic Kandiudult).
In October 1993 and 1994, after the cotton stalks were
shredded, half the plots were seeded with rye cover (110
lb of seed/acre). Plots were seeded in 7.5-in. rows using a
John Deere 750 grain drill.
In a separate operation immediately prior to planting,
half the subsubplots were subsoiled using a KMC four-
row subsoiler within 6 in. of the previous year’s rows. In
mid-May, cotton (‘DES 119’) was seeded within 6 in. of
the previous year’s rows with a four-row Case-IH 900
series planter equipped with Yetter wavy coulters. We at-
tempted to maintain the same wheel tracks and rows from
year to year. However, because the old rows were no longer
visible, locating wheel tracks was more difficult in the
disked than in the non-disked plots
Nitrogen (80 lb N/acre as ammonium nitrate) was ap-
plied in a split application, half at planting and half one
month after planting. For each application, N was banded
approximately 4 in. deep and 6 in. from the rows. Lime, P,
K, S, B and Mn were applied based on soil test results and
Clemson University Extension recommendations. Weeds
were controlled with a combination of herbicides, cultiva-
tion (disked plots only) and hand-weeding. Insects were
controlled by applying aldicarb (0.75 lb ai/acre) in-fur-
row. Other insecticides were applied as needed.
Soil strength was measured in early June with a 0.5-in.-
diameter, 30o solid angle cone tip, hand-operated, record-
ing penetrometer (Carter, 1967). Strength measurements
were recorded to a depth of 24 in. at nine positions across
a mid-plot row (from non-traffic midrow to traffic
midrow). Each measurement was the mean of three probings
from each subsubplot. Data were recorded on index cards
and digitized into the computer using the method described
by Busscher et al. (1986). Data were log transformed be-
fore analysis for normalization (Cassel and Nelson, 1979).
Along with the cone indices, water contents were mea-
sured at 4-in. depth increments in the non-wheel-track
midrow and in the row. These selected water contents
were considered representative of the water contents for
each subsubplot.
In mid to late October, cotton was chemically defoli-
ated. In early November, seed cotton yield was measured
by harvesting two interior rows with a two-row spindle
picker. Each harvest bag was subsampled, and the subsample
was saw-ginned to measure lint percent. Seed cotton yield
was multiplied by lint percent to estimate lint yield.
Data were analyzed using ANOVA and the LSD mean
separation procedure (SAS Institute Inc., 1990). Unless
otherwise specified, differences were significant at P =
In late summer 1994, hail ruined part of the field that
included half of replicate one. After this, the replicate was
ignored and the other three were used for analysis.
For both years and over all tillage treatments, cone
index differed with depth (Table 1 and Fig. 1). The highest
cone indices were found at the 12- to 16-in. depths, the
bottom of the E horizon. This high subsoil strength was
the main reason for implementing the deep tillage.
Some cone index differences with depth were caused
by water content changes (Table 1). For example, the softer
soil below the hard layer (> 16 in.) was also wetter. At
this depth, soil type generally changed from loamy sand to
sandy clay loam. The sandy clay loam held more water and
had structure. The higher water content reduced cone in-
dex and provided nourishment for the root, if it could
penetrate the pan above. The structural faces provided zones
of weakness along which roots could grow, even if the
soil dried and hardened.
Cone index varied with position across the row (Table
2 and Fig. 1). These differences distinguished lower
strength under the non-wheel-track midrow (Fig. 1, posi-
tion = 0 in.) than the wheel-track midrow (position = 38
in.). The lowest cone indices were found in the midrows
(position = 19 in.) because of this year’s deep tillage or
residual effects from past deep tillage in the non-deep-
tilled treatments.
Mean profile cone indices (M) did not differ between
disked and non-disked treatments. An exception to this
was the 1994 non-deep-tilled treatments where disked
treatments had lower M (Table 3). This was a result of
lower cone indices in the surface 4 in., caused by the
disking. This zone of lower strength was apparent in the
other cases (Fig. 1) but not significantly different.
As expected, M for the deep-tilled treatment was lower
than for the non-deep-tilled treatment (Table 3). An ex-
ception to this was the disked treatment in 1994 where
M’s were about the same for both deep tilled and non-
deep-tilled treatments. The similarity of the M’s could be
explained partly by the residual effects of 1993 subsoiling
in the non-deep-tilled treatment, giving this profile a loos-
ening pattern similar to the deep-tilled treatment (Fig. 1).
Also, since both treatments were disked, the upper parts
of both profiles were loosened.
Most strength interactions with cover were accompa-
nied by water content differences. The higher strengths
had lower water contents. Most of these differences were
in the lower half of the measured profile.
In the non-disked treatments, the rye cover treatment
had lower cone indices (and higher water content) than the
fallow treatment (Table 4). This would be consistent with
better infiltration usually associated with treatments that
have better cover.
The opposite was seen in the disked treatments, where
the fallow treatment had the lower cone indices (and higher
water contents). This would be consistent with root uptake
by the rye.
In 1994, cotton yield was higher for fallow cover in the
non-disked treatments and for rye cover in the disked treat-
ments (Table 4). This was a result of the large amount of
cover in the 1994 rye cover treatments that made planting
difficult in the non-disked rye cover and added a signifi-
cant amount of organic matter to the disked treatment
(Bauer et al. 1995).
In 1995, in the non-subsoiled treatments, cone indices
were lower for the non-disked rye than fallow and higher
for the disked rye than fallow (Table 4). Lower cone indi-
ces for the non-disked rye suggested that the cover (and
the roots from the cover crop growing within the profile)
helped maintain low strengths, even for soils with hard-
pans at 12- to 16-in. depth. Higher strengths for the disked
rye suggest that disking can eliminate these reductions in
strength. Since the profile as a whole was higher in strength
and since disking loosened the upper part of the profile
(as seen above), the lower part of the profile, the pan,
would have had to be compacted. Lower cone indices sug-
gest higher yields for the non-disked treatment. Higher
yields were found, although they were not significantly
different. Also not significantly different, the 1994 cone
index data showed the same trend as the non-subsoiled
1995 cone index data. Water contents for these treat-
ments were not significantly different.
Cover crops have a number of known advantages: re-
ducing erosion, reducing leaching of nutrients and increas-
ing organic matter. It is also advantageous to know that
they can be used without reducing cotton yield (and per-
haps increasing it) by helping maintain low soil strength.
We thank E.E. Strickland and B.J. Fisher for technical
support. Mention of trademark, proprietary product or ven-
dor does not constitute a guarantee or warranty of the
product by the U.S. Department of Agriculture and does
not imply its approval to the exclusion of other products
or vendors that may also be suitable.
Bauer, P.J., and W.J. Busscher. 1993. Effect of winter cover on soil
moisture content in conventional and strip tillage cotton. pp. 8-10.
In: P.K. Bollich (ed.). Proc. So. Conserv. Till. Conf. for Sustain.
Agric., Monroe, Louisiana.
Bauer, P.J., W.J. Busscher and J.M. Bradow. 1995. Cotton response
to reduced tillage and covercrops in the southeastern Coastal
Plain. pp. 100-102. In: W.L. Kingery and N. Buehring (eds.).
Proc. So. Conserv. Till. Conf. for Sustain. Agric., Jackson,
Busscher, W.J., R.E. Sojka, E.J. Sadler and C.W. Doty. 1986.
Simplified data analysis for an inexpensive manual analogue
penetrometer. Comput. and Electron. in Agric. 1:197-204.
Carter, L.M. 1967. Portable penetrometer measures soil strength
profiles. Agric. Eng. 48:348-349.
Cassel, D.K., and Nelson, L.A. 1979. Variability of mechanical
impedance in a tilled one-hectare field of Norfolk sandy loam.
Soil Sci. Soc. Am. J. 43: 450-455.
Langdale, G.W., R.L. Wilson, Jr. and R.R. Bruce. 1990. Cropping
frequencies to sustain long-term conservation tillage systems.
Soil Sci. Soc. Am. J. 54:193-198.
SAS Institute, 1990. SAS Language: Reference, Version 6. SAS
Institute Inc., SAS Circle, Box 8000, Cary, NC 27512-8000
Table 1. Cone indices and water contents by depth.
Cone Index (Atm) Water Content (lb/100 lb)
Depth (in.) 1994 1995 1994 1995
2 10.3f* 8.9f 5.8e 10.6c
6 21.7e 18.6e 6.0de 10.0d
10 36.1d 24.5d 6.8c 10.0d
14 57.1a 38.5a 6.6cd 10.2cd
18 46.0b 30.3c 8.3b 11.6b
22 41.6c 31.3b 10.3a 12.9a
* Means by year with the same letter are not different (LSD at 5%).
Table 2. Cone indices by position across the row.
Cone Index (Atm)
Position 1994 1995
Non-wheel track 24.3b 19.6b
In row 19.7c 11.9c
Wheel track 31.2a 22.3a
* Means by year with the same letter are not different (LSD at 5%).
Table 3. Mean profile cone index by tillage treatment.
Tillage Cone Index (Atm)
Surface Deep 1994 1995
Non-disked Non-subsoiled 27.4a 21.4a
Non-disked Subsoiled 23.1b 17.5b
Disked Non-subsoiled 23.3b 20.7a
Disked Subsoiled 22.6b 17.5b
* Means by year with the same letter are not different (LSD at 5%).
Table 4. Mean profile cone index and yield by deep tillage,
surface tillage and cover.
Surface Deep Cone Index (Atm) Yield (lb/acre)
Tillage Tillage Cover 1994 1995 1994 1995
Disked Subsoiled Fallow 21.2 16.1b 1060 665
Rye 24.0 16.8b 1200 724
subsoiled Fallow 22.3 18.3b 1110 695
Rye 24.3 21.2a 1210 619
disked Subsoiled Fallow 23.4 16.9b 1299 567
Rye 22.8 16.2b 1010 724
subsoiled Fallow 29.0 21.2a 1240 624
Rye 25.9 19.6b 1000 838
* Means by year with the same letter are not different (LSD at 8%).
Fig. 1. Isostrength lines for treatment profiles in spring 1994
averaged over covers.
T.A. Castillo, T.K. Keisling and L.R. Oliver1
Along-term field study was initiated in 1996 to
evaluate tillage methods and herbicide treatments
for redvine control in soybeans (Glycine max).
Aerial photography and Global Positioning System (GPS)/
Global Information Systems (GIS) were used to monitor
redvine movement. At trial initiation, redvine populations
averaged 15 to 25 per m2 and resulted in 42 to 50%
groundcover. A split plot design was used with tillage type
as the main plot and herbicide treatment as the subplot.
Tillage types included no-till, conventional, hyperbolic
subsoiler and moldboard plow. Subsoiling and plowing op-
erations were conducted in the fall of 1996. Herbicide
treatments included no herbicide, glyphosate at 1.1 kg ai/
ha (1.0 lb ai/acre) applied annually to V2 and V6 soybeans
and dicamba at 2.2 kg ai/h (2.0 lb ai/acre) applied 2 weeks
prior to 1996 soybean harvest. When a herbicide was not
used, moldboard plowing was the only tillage type that
provided acceptable season-long control (83%). The
subsoiler provided 50% control of redvine, but by harvest
regrowth had occurred, resulting in only 24% control. Stem
counts were reduced by moldboard plowing and subsoiling.
Conventional tillage actually increased stem counts.
Glyphosate increased control of redvine for all tillage treat-
ments except moldboard plowing. Glyphosate at V2 and
repeated at V6 provided redvine control for one month
after the V6 treatment; however, late-season regrowth re-
sulted in only 54 to 66 % control at harvest. Dicamba
provided 96% control regardless of tillage type. Redvine
density did not affect soybean yield in 1997.
As reduced tillage systems become more popular,
redvine and other perennial weeds are becoming an in-
creasing problem in the Mississippi Delta (Elmore, 1984).
Redvine has an extensive underground stem and root sys-
tem, capable of vegetative propagation (DeFelice and
Oliver, 1980). Control of this weed requires that a sub-
stantial concentration of herbicide reach the root system
(Shaw and Mack, 1991). If applied during the fall, when
the redvine plants are translocating sugars to their root
structures, dicamba can reduce groundcover levels for at
least two years (Elkins et al., 1996). Disruption of the
root structure by deep tillage has also been found to re-
1Department of Agronomy, University of Arkansas, Fayetteville, AR 72701.
duce redvine groundcover levels (Elkins et al., 1996). Till-
age operations may also contribute to the spread of peren-
nial weeds throughout a field (Soteres and Murray, 1982).
The objective of this study was to further develop redvine
control programs in Roundup Ready soybeans with tillage
methods and systemic herbicides and to monitor the re-
growth and movement of redvine within the treatment.
A 10-ha farmer-cooperator field near Keiser, Arkan-
sas, containing a high natural population of redvine was
selected for study. A split plot design with four replica-
tions was used. The main plots consisted of four tillage
methods: no-till, conventional tillage, hyperbolic subsoiler
and moldboard plow. Subsoiling and moldboard plowing
operations were conducted upon initiation of the experi-
ment in the fall of 1996. Subplots were herbicide treat-
ments and included dicamba applied two weeks prior to
harvest in 1996 at 2.2 kg ai/ha, glyphosate applied annu-
ally to V2 and V6 soybeans at 1.1 kg ai/ha and an untreated
check. ‘Asgrow 4701RR’ soybean cultivar was drill seeded
to the 15- x 15-m plots 13 May 1997. Visual control
ratings were taken at planting, one, two and three months
after planting and at harvest. Redvine stem counts/m2 were
also taken from the same plot area each year prior to
harvest. The entire plot area was harvested for soybean
yield. Original plot locations were mapped with Global
Positioning Systems (GPS) technology, and aerial photo-
graphs are being taken semiannually to monitor the loca-
tion and movement of redvine with the use of Geographic
Information Systems (GIS) software. All data were sub-
jected to analysis of variance, with means separated by
Fishers Least Significant Difference (LSD) at the 0.05
significance level.
Tillage Alone
When no herbicide was used for redvine control, mold-
board plowing was the only tillage treatment that provided
acceptable control for the entire growing season (Fig. 1).
When the top portion of the soil profile was turned, sub-
terranean redvine parts were sliced off 20 cm below the
soil surface. Regrowth from the remaining taproot was
hindered and may have required the formation of new buds
from root tissue. Fragmented stem segments were depos-
ited at the soil surface. Exposure to cold and wet condi-
tions during the winter of 1996-1997 desiccated these Shaw, D.R., and R.E. Mack. 1991. Application timing of herbicides for
fragments and prevented regeneration. Both factors led to the control of redvine (Brunnichia ovata). Weed Technol.
an 83% reduction in stem counts (Table 1). Control with 5:125-129.
the hyperbolic subsoiler was much less. The subsoiler dis- Soteres, J.K., and D.S. Murray. 1982. Root distribution and
reproductive biology of honeyvine milkweed (Cynanchum
turbed less than half of the soil matrix, leaving many es- laeve). Weed Sci. 30:158-163.
tablished roots and rhizomes intact for regrowth. At har-
vest, control with the subsoiler was similar to that with
conventional tillage but higher than the no-till check (Fig. Table 1. Reduction in redvine stems/m2 1996-1997.
1). Only the conventional-tillage method increased stem Herbicide program
counts (Table 1). Tillage Level Untreated Glyphosate* Dicamba**
Tillage + Glyphosate No-till 11 21 96
Sequential applications of glyphosate increased redvine Conventional -25 19 99
control over that of tillage alone, except for moldboard Subsoiler 38 46 100
plowing (Fig. 2). Glyphosate provided control for one Moldboard 83 72 100
month after treatment; however, late summer regrowth LSD (0.05%) = 22
*Glyphosate at 1.1 kg ai/ha applied V2 and V6
caused final ratings to decline, resulting in 54 to 66% **Dicamba at 2.2 kg ai/ha applied preharvest 1996.
control for all tillage types. Glyphosate reduced stem
counts only in the conventional tillage plots (Table 1).
Tillage + Dicamba
Regardless of tillage type, dicamba provided excellent 100
control for the entire year (Fig. 3). Only minimal regrowth
occurred late in the season. 80
Soybean Yield
Redvine density did not affect yield. While the pres- 60
ence of redvine may alter the microclimate through com- 40
petition for light and soil moisture, the less-than-com-
plete plot coverage and narrow-row soybeans compensated 20
for the interference. Although redvine may not directly
affect returns, the long vines often entangle machinery, 0
causing substantial tillage and harvest complications. 012345
No-till Conv Subsoil Moldboard
CONCLUSIONS Months After Planting
Acceptable redvine control requires that the underground Fig. 1. Redvine control with tillage alone (no herbicide), 1997.
portion of the plant be killed by either moldboard plowing
or the use of dicamba. Split applications of glyphosate can
keep redvine at a manageable level below the crop canopy.
Subsoiling provided early-season control, but stem counts
at harvest were not reduced over no-till. Conventional till- 100
age may actually increase redvine populations and areas of
infestation. Redvine did not affect soybean yields. 80
The authors would like to thank the Arkansas Soybean
Promotion Board for funding this project and the Jimmy 40
Flecher Farm for providing the test area. 20
DeFelice, M.S., and L.R. Oliver. 1980. Redvine and trumpetcreeper 0
No-till Conv Subsoil Moldboard
control in soybeans and grain sorghum. Ark. Farm Res. 29(3):5. 0 1 2 3 4 5
Elkins, W.C., D.R. Shaw, J.D. Byrd, Jr., M.A. Blaine and C.H. Tingle. Months After Planting
1996. Long-term redvine control. Proc. South. Weed Sci. Soc.
49:22. Fig. 2. Redvine control with tillage and glyphosate (1.1 kg ai/
Elmore, C.D. 1984. Perennial vines in the Delta of Mississippi. Miss. ha) applied to V2 and V6 soybeans (1 and 2 months after
Agric. and For. Exp. Stn. Bull. 927. planting in 1997).
% Control % Control
0 0 1 2 3 4 5
No-till Conv Subsoil Moldboard
Months After Planting
Fig. 3. Redvine control with tillage and dicamba (2.2 kg ai/ha)
applied in fall 1996.
% Control
M.B. Daniels, S.L. Chapman, R. Matlock and A. Winfrey
Underlying soil fertility problems such as high so-
dium levels, excess soluble salts and micronutri-
ent imbalances can limit plant response to nitro-
gen, phosphorus and potassium fertilizers and lime even
when soil test recommendations warrant such additions.
Management options for these soils are sometimes lim-
ited due to practical and economic constraints. The objec-
tive of this study was to determine if the use of precision
agricultural technology could provide information that
would increase fertility management options on problem
Grid soil sampling is primarily being used as a basis
for variable rate application of fertilizers and lime. Re-
gardless of variable rate fertilizer technology, grid soil
sampling may be an important management tool. It pro-
vides information at a level of detail that may be neces-
sary for other purposes, such as setting realistic yield
goals, explaining yield variability and trouble shooting
problem soils.
Plant response can vary within a field with problem
soils ranging from seedling death in some locations to
normal growth and yield at other locations. This variabil-
ity can make it difficult to diagnose and remedy the prob-
lem with normal composite soil sampling from good and
bad areas. Intensive soil sampling may provide informa-
tion so that the problem can be adequately identified and
the spatial extent of the problem adequately delineated.
Ultimately, this increased knowledge may lead to increased
management strategies for problem soils.
The study was conducted in the spring of 1997 in south-
western Hot Spring County in a 70-acre production field.
Historically, soybean yields in parts of this field have been
severely limited due to excess soluble salts. Within this
field, the soils are mapped as Adaton, Gurdon and Sardis
silt loams. The Gurdon series is closely related in texture
and landscape position to the Foley silt loam, which is
characterized by a natric (high sodium content) horizon.
In order to determine the distribution of soluble salts
and sodium within the field, soil samples were obtained
on approximately a 2.5-acre grid while the field was fal-
low. The grid points were somewhat irregular (Fig. 1) and
more dense where there was visual evidence of salt prob-
lems (lack of vegetation) to ensure that problems areas
smaller than 2.5 acres were not excluded from the sam-
pling. At each grid point, samples were collected with an
NRCS probe truck using a 3-in.-diameter collection tube.
Samples were taken from four depths down to 24 in. in 6-
in. increments. The samples were shipped to the Univer-
sity of Arkansas Soil Test Lab at Marianna for routine soil
The latitude and longitude coordinates were determined
for each grid point with a hand-held DGPS (Post Process-
ing). Coordinates for the perimeter of the field were also
recorded. Soil nutrient maps were constructed using
SSToolbox GIS software (SST Development Group, Inc.).
Soil test point data was converted to surface data using
kriging procedures.
Soil test results indicated low fertility levels of P, K
and pH (Table 1). Field averages of electrical conductivity
(EC) and sodium did not indicate excessive levels of
soluble salts or sodium at any depth interval. However,
sodium levels at all depth intervals were highly variable
ranging from 100 lb/acre to greater than 999 lb/acre
(Maximum value reported by lab) with coefficients of varia-
tion, ranging from 62 to 80%. For a silt loam texture, it is
thought that sodium values exceeding 500 lb/acre would
adversely impact crop growth. The number of acres ex-
ceeding this threshold value increased from 6 acres in the
top 12 in. to 7 acres at the 12- to 18-in. depth interval to
24 acres at the 18- to 24-in. depth interval (Fig. 1 and 2).
Because the farmer was considering land leveling this
field, elevation data (locations recorded with DGPS) rela-
tive to a benchmark datum was obtained from Bowls Sur-
veying (Fig. 3). Overlaying procedures using GIS software
were performed on the maps in Fig. 2 and 3 to determine
if land leveling would expose more acreage exceeding the
500-lb/acre sodium threshold (Fig. 4). From this analysis,
it was determined that potentially 4 more acres of sodium
exceeding the threshold might occur in the top 12 in. if
land leveling was performed.
1Published in Arkansas Soil Fertility Studies 1997, Wayne E. Sabbe, editor.
Arkansas Agricultural Experiment Station Research Series 459:24-28.
The results obtained from this study have been used to
help make crucial management decisions related to this
field. From Fig. 1, it was determined that 8% of the field
could suffer crop damage from salt. From Fig. 2, 3 and 4,
it was determined that land leveling could potentially in-
crease the sodium hazard in the top 12 in. of the root zone
by 4 acres up to a total of 13% of the acreage. The farmer
proceeded with land leveling because he felt the advantage
of better water management outweighed the small increase
(5%) in sodium hazard.
By knowing the sodium distribution, the producer was
able to prioritize his management options. Instead of fo-
cusing his attention on the 8% of the field affected by
sodium, he can address the low fertility problems in the
other 92% of the field where pH, P and K are limiting
crop production. Before, it was assumed that poor crop
production from the field as a whole was a result of high
salt levels rather than poor fertility.
SSToolbox. 1996. SST Development Group, Inc. 824 N. Country Club
Rd., Stillwater Oklahoma 74075-0918.
Table 1. Selected soil test results by depth.
Depth ph P K Na EC
In. --------lb/acre-------- µmhos/cm
0-6 Mean 4.7 11 67 320 190
s.d. (+/-) 0.3 4 13 253 265
Minimum 3.9 10 50 100 35
Maximum 5.6 29 105 999 1366
6-12 Mean 4.8 11 52 328 128
s.d. (+/-) 0.5 4 8 220 140
Minimum 3.9 10 50 113 24
Maximum 6.8 34 105 999 620
12-18 Mean 4.7 11 53 350 134
s.d. (+/-) 0.4 2 12 219 141
Minimum 3.9 10 50 143 24
Maximum 6.8 19 129 999 620
18-24 Mean 4.6 10 57 418 153
s.d. (+/-) 0.3 15 269 148
Minimum 4.0 50 136 31
Maximum 6.2 148 999 682
Fig. 1. Map of field boundary, soil sample location and sodium
(lb/acre) distribution in the top 6 in. Grid cells represent
10,000 ft2 (~0.25 acres).
Fig. 2. Map of sodium (lb/acre) distribution at 18 to 24 in. Each
grid cell represents 10,000 ft2 (~0.25 acres).
Fig. 3. Map of cut sheet used for land leveling. Positive values
refer to areas of fill (ft) while negative values refer to areas of Fig. 4. Map of intersection between cut areas and sodium
removal (ft). Data furnished by Bowls Surveying, England, distribution (>500 lb/acre) at 18 to 24 in. Map created by using
Arkansas. overlay techniques on Fig. 2 and 3.
Ernest H. Flint, Jr., Glover B. Triplett, Jr., Seth M. Dabney,
William H. Batson, Dawn S. Luthe and Clarence E. Watson1
Performance of no-tillage cotton (Gossypium
hirsutum L.) in the mid-South has ranged from yield
decreases (Brown et al., 1985; Stevens et al., 1992)
to yield increases (Bradley, 1995; Triplett et al., 1996).
Both the Brown et al. (1985) and Stevens et al. (1992)
studies were conducted for three years with no-tillage
yields improving as studies progressed. Triplett et al.
(1996) reported reduced no-tillage yields for the first year
of their study with improved productivity as time pro-
gressed so that no-tillage yields were greater than conven-
tional during years three through five. Thus, a period of
time may be required for cotton yields to reach their full
potential following implementation of no-tillage practices.
Site characteristics may be a factor, as well, in perfor-
mance of different systems as all studies cited were lo-
cated on coarse ormedium textured soils.
In the non-irrigated study reported by Triplett et al.
(1996), percentage yield improvement with no-tillage was
greatest during moderately dry years. This implies that
no-tillage improved moisture relations in some manner.
Increased moisture for the crop could have resulted from
increased rainfall infiltration through established macro-
pores, slower runoff due to mulch, reduced evaporation
under mulch, some factor not yet identified or a combina-
tion of factors. With a pattern of improved crop produc-
tivity clearly established for no-tillage in longer-term stud-
ies for cotton as well as other crops (Bruce et al., 1995),
efforts to identify mechanisms involved become appro-
priate. An area that has received scant attention in no-
tillage cotton research is the possible contribution of my-
corrhizae to the growth and productivity of the crop.
In mycorrhizal associations, fungi of the family
Endogenaceae colonize roots of host plants. Most plant
families form mycorrhizal associations, including cotton,
corn (Zea mays L.), wheat (Triticum aestivum L.) and
many weed species present between crops or concurrent
with the crop. In these associations the hyphae of the fun-
gal species invade plant roots and form arbuscules, which
facilitate ready exchange of nutrients between the host
and fungus, resulting in the association known as VAM
(Vesicular Arbuscular Mycorrhizae). This association can
be parasitic, benign or beneficial, but it is commonly mu-
1Area Extension Agent, Prof. Plant and Soil Sci. Dept., Agronomist USDA-
ARS National Sedimentation Lab., Prof. Plant Path., Prof. Biochemistry and
Head MAFES Experimental Statistics.
tualistic with the fungus receiving energy from the plant.
The plant, in turn, may receive several benefits from the
association. Rich and Bird (1974) reported that early-sea-
son root and shoot growth of cotton was increased in the
presence of mycorrhizal fungi and that these plants flow-
ered and matured bolls earlier. Zak et al. (1998) suggest
that the fungus forms a hyphal network in the soil that can
serve as an extension of the plant root system. Thus, a
seedling that is colonized early can explore a much greater
soil volume than is possible with a newly developing root
system. Inorganic ions such as P and Zn are absorbed by
the fungus and transferred to the plant. This improvement
of P nutrition is a critical factor in soils with low P con-
tent. In turn, this can lead to reduced fertilizer require-
ments and more efficient use of soil nutrients (Marschner
and Dell, 1994).
The hyphal network may also transport moisture to the
plant, replacing water lost through transpiration and better
maintaining plant turgor during dry periods. Mycorrhizal
plants recover faster following moderate water deficits
(Safir et al., 1971). This also implies that VAM plants
may exhaust stored soil moisture more thoroughly than
plants without an extensive hyphal network in place. The
colonized plants may also avoid some stresses caused by
nematodes (Hussey and Roncadori, 1982) and some plant
diseases (Linderman, 1992). Tillage fragments the hyphal
network so that it must be reestablished as the crop devel-
ops. With no-tillage, an existing network remains intact
and may be exploited by seedling plants (Zak et al., 1998).
The study reported here was initiated to investigate differ-
ences in cotton growth, nutrient uptake and VAM coloni-
zation as influenced by tillage practices.
No-tillage following a killed wheat cover crop and con-
ventional tillage cotton plots established in 1988, as de-
scribed by Triplett et al. (1996), were used in these stud-
ies. The cotton was planted in early May 1996. The treat-
ments described below were imposed on individual plots
and/or plants within the study area.
Plant Development
Node counts and plant height measurements were be-
gun 5 June when plants were at the four-node stage and
approximately 5 in. tall. Measurements were continued on
a weekly basis until 6 July.
Root Colonization
Root tissue samples were selected at random from both
tillage treatments in two blocks. Block A had a depth to
fragipan of 34 in., a 3 to 4% slope and a history of equal
yields for both tillage systems. Block B had a 5 to 6%
slope, a fragipan depth of 22 in. and a yield history of no-
tillage greater than conventional. Plants were sampled on
29 June at the 10-node stage. Five 1-cm sections of root
tissue were selected from each of four plants in each
tillage system. Root segments were stained, and coloniza-
tion sites per cm of root length were recorded.
Hyphal Network and Phosphorus Uptake Studies
Three days after emergence, the following treatments
were imposed on 10 individual seedlings in both tillage
blocks: 1) no disturbance, 2) a 4-in.-diameter core cutter
used to cut around the plant and a 6-in.-deep core re-
moved, wrapped in nylon mesh with 60µ diameter open-
ings and replaced and 3) core cut as in 2) but not removed.
The nylon mesh openings were small enough to exclude
roots but permitted hyphal penetration. To assess the hy-
phal network, plants were allowed to develop until mature
with open bolls. The fabric was then removed, stained and
examined for mycorrhizal hyphae. Counts of a single fab-
ric sample from each plant were made within a 1000µ
microscope reticle scale, rotating the eyepiece to create a
circle of 1000µ. Each hyphal strand crossing a fabric pore
was counted and recorded.
In the phosphorus uptake study, 10 days after emer-
gence one microcurie of 32P orthophosphate was injected
1 in. deep, 6 in. from individual cotton seedlings in treat-
ments one, two and three described above. At the initial
sampling, plants had only one fully formed leaf. This in-
creased to two by the last sampling. Leaves from four
plants were sampled one, four and eight days after 32P
application by cutting four 1-cm-diameter discs from tis-
sue of each leaf. The amount of radioactivity taken up by
the leaves was determined by scintillation spectroscopy.
Physiological Evaluations
These studies were done with a portable Li-Cor LI-
6400 Photosynthesis System through courtesy of the MSU
Crop Simulation Laboratory. The data were collected on
13 August 1996 under clear skies with temperatures in
the range of 89 to 91 degrees F. Data collected included
evaluations of stomatal conductivity, transpiration and level
of photosynthesis.
In preliminary results from these studies, the mean node
number for conventional tillage and no-tillage plants were
similar (4.2 and 4.3, respectively). Initial plant heights
were significantly different (5.0 vs. 5.8 in., respectively)
for till and no-tillage. During the measurement period,
no-tillage plants developed a node each 4.4 days vs. 4.7
days for plants in tilled soil. Plants in tilled soil grew
significantly more slowly (0.54 in./day) than no-tillage
plants (0.83 in./day). Although seedlings emerged in both
tillage systems at the same time, plants in the no-tillage
treatment grew taller and developed more rapidly than those
in the tilled area. Vivekanandan and Fixen (1991) reported
a similar vegetative growth response in corn which they
attributed to mycorrhizal activity.
In the colonization study, the overall VAM coloniza-
tion intensity was greater for no-tillage in the deeper soil
(Table 1). However, the colonization pattern shown here
does not explain the previously observed crop yield pat-
tern of equal yields for both tillage systems in area A.
Little information is available to indicate how degree of
colonization influences mycorrhizal symbiosis.
In the hyphal network study, 34 hyphae/1000µ circle
crossed the nylon mesh barrier with no-tillage. This was
significantly greater than the 9 hyphae/1000µ circle in the
tilled treatment. By the time the mesh and plants were
removed, the plant root system completely occupied the
confines of the mesh cylinder. The greater hyphal counts
for no-tillage indicate that the hyphal strands were more
numerous in the untilled soil, complementing the greater
colonization intensity shown in Table 1. This supports, but
does not confirm, the presence of a more established hy-
phal network in untilled soil.
In the phosphorus uptake study, no radioisotope activ-
ity level significantly greater than background was detected
until eight days following injection of the tracer and then
only for the uncut treatment (Table 2). Since P is immo-
bile in the soil, the isotope was accessed by the plant
either by root uptake or transported through VAM hyphae.
Lack of uptake for the cut treatment supports the premise
that the hyphal network was disrupted by cutting and was
not reestablished and functional when the small plants were
Results from the physiological measurements are shown
in Table 3. The no-tillage cotton plants were more ac-
tively transpiring at the time measurements were taken.
This suggests that plants under no-tillage were able to
obtain more moisture from the soil than under conven-
tional tillage; however, the level of photosynthesis was
similar for the two tillage treatments.
Results from the studies with cotton reported here com-
pare favorably with published reports dealing with VAM
and other crops. While no cause-and-effect relationships
are definitely established, evidence is such that the role of
VAM in no-tillage cotton production warrants further ex-
Bradley, J.F. 1995. Success with no-till cotton. pp. In: M.R.
McClelland et al. (ed.). Conservation -tillage systems for cotton.
Ark. Agric. Exp. Stn. Spec. Rep. 169:31-35.
Brown, S.M., T. Whitewell, J.T. Touchton and C.H. Burmester. 1985.
Conservation tillage systems for cotton production. Soil Sci. Soc.
Am. J. 49:1256-1260.
Bruce, R.R., G.W. Langdale, L.T. West and W.P. Miller. 1995.
Surface soil degradation and soil productivity restoration and
maintenance. Soil Sci. Soc. Am. J. 59:654-660.
Hussey, R.S., and R.W. Roncadori. 1982. Vesicular arbuscular
mycorrhizae may limit nematode activity and improve plant
growth. Plant Dis. 66:9-14.
Linderman, R.E., 1992. Vesicular-arbuscular mycorrhizal and soil
microbial interactions. In: Mycorrhizae in sustainable
agriculture. Am. Soc. Agron. Spec. Pub 54:45-69.
Marschner, H., and B. Dell. 1994. Nutrient uptake in mycorrhizal
symbiosis. Plant and Soil. 159:89-102.
Rich, J.R., and G.W. Bird. 1974. Association of early-season
vesicular-arbuscular mycorrhizae with increased growth and
development of cotton. Phytopathology 64:1421-1425.
Safir, G.R., J.S. Boyer and J.W. Gerdemann. 1971. Mycorrhizal
enhancement of water transport in soybeans. Science 172:581-
Stevens, W.E., J.R. Johnson, J.J. Varco and J. Parkman. 1992. Tillage
and winter cover management effects on fruiting and yield of
cotton. J. Prod. Agric. 5:570-575.
Tinker, P.B.H. 1975. Effects of vesicular mycorrhizae on higher
plants. Phil. Trans. R. Soc. Lond. B. 273:445-461.
Triplett, G.B., Jr., S.M. Dabney and J.H. Siefker. 1996. Tillage
systems for cotton on silty upland soils. Agron. J. 88:507-512.
Vivekanandan, M., and P.E. Fixen. 1991. Cropping systems effects on
mycorrhizal colonization, early growth, and phosphorus uptake of
corn. Soil Sci. Soc. Am. J. 55: 136-140.
Zak, J.C., B. McMichael, S. Dhillion and C. Friese. 1998. Arbuscular-
mycorrhizal colonization dynamics of cotton (Gossypium
hirsutum L). growing under several production systems on the
Southern High Plains, Texas. Agriculture Ecosystems &
Environment. (In press)
Table 1. Colonization of cotton roots by mycorrhizal fungi
under no-tillage and conventional-tillage culture.
Tillage Method
Area No-tillage Conventional
-------------sites per/cm root-------------
A 12.25a1 3.10c
B 6.35b 4.90bc
Mean 9.3 4.0
LSD (0.05) = 2.37
1Means not followed by the same letter are different at the 0.05 level of
Table 2. Relative amount of 32P (CPM) found in leaf tissue.
Treatment Mean
Uncut 17.43a1
Cut 3.73b
Mesh 2.53b
1Means not followed by the same letter are different at the 0.05 level of
Table 3. Evaluations of stomatal conductivity, transpiration
and level of photosynthesis.
Treatment Mean
Stomatal Conductivity
No-Tillage 0.219a1
Conventional Tillage 0.171b
LSD(0.05)= 0.04
No-Tillage 3.77a
Conventional Tillage 3.18b
LSD(0.05)= 0.57
No-Tillage 18.85a
Conventional Tillage 17.66a
1Means not followed by the same letter are different at the 0.05 level of
R.N. Gallaher and R. McSorley1
Sweet corn (Zea mays L.) is an economically impor-
tant crop for Florida. The hot-humid climate in
Florida provides an environment for off-season
sweet corn production at a time when most of the U.S. is
too cold for corn growth. This same environment also is
favorable for large populations of insect pests, which can
reduce yield and quality. Past studies have shown that the
use of the insecticides Counter (terbufos) and Furadan
(carbofuran), at planting of field corn, can significantly
increase yield (Gallaher, 1983, 1986a,b; Gallaher and
Baldwin, 1985; Espaillat and Gallaher, 1989). All of the
above research in the 1980’s was with the use of Furadan
15G. This granular formulation was widely used at the
time but became restricted and largely unavailable and was
replaced with a non-granular formulation. The granular prod-
uct had the advantage of ease of application and incorpo-
ration in the seed furrow or row and was easily activated
around the seed zone. The liquid product, Furadan 4F avail-
able for use at present in Florida, is thought to require
more sophisticated equipment in order to obtain good ac-
tivation in the seed furrow-zone.
In these earlier studies with field corn, we found that
Furadan performed better than Counter under no-tillage
management, but the two products were equally effective
in conventional tillage systems. Another discovery was
that field corn hybrids responded more favorably to the
insecticide that had been used in the hybrid breeding pro-
gram. It was not unusual to obtain 40 to 50 bu/acre yield
increases from the use of insecticides applied in the row
at planting time (Espaillat and Gallaher, 1989). These ma-
terials also show activity as nematicides (Norton et al.,
1978). After the loss of the granular formulation of
Furadan in Florida, sales of this product were significantly
The objectives of this investigation were to determine
1) the yield differences among five sweet corn hybrids
under no-till management, 2) the effectiveness of the use
of Furadan 4F formulation sprayed in a band over the corn
row at planting and 3) effects on plant-parasitic nematode
1Agronomy Department and Entomology and Nematology Department
respectively, University of Florida, Institute of Food and Agricultural Sciences,
Gainesville, FL.
The split-plot experiment was conducted on a Arredondo
fine sand on the University of Florida, Green Acres
Agronomy Field Laboratory in 1997. Main plots were five
sweet corn hybrids (‘XPH 3084’; ‘VXT 5 Forever’; ‘VNE
2 Endeavor’; ‘VNT 5 Punchline’; ‘XPH 3105’), planted at
28,000 plants/acre, in four-row plots, 2.5 ft wide and 20
ft long. The two subplots were with the application of
carbofuran (formulated as Furadan 4F) at 1.0 lb ai/acre
(the labeled rate) versus a control without application of
The experimental site was planted to a cover crop of
‘Tift Blue” lupin (Lupinus angustifoilus L.) in the fall of
1996. On 17 April 1997, the sweet corn was planted di-
rectly into the standing lupin with a Brown-Harden In-
Row Subsoil (Strip-till) no-tillage planter, using John Deere
Flexie 71 planter units. On 21 April 1997, 1.8 quarts Bicep
II (mixture of atrazine and metolachlor)/acre plus 2 quarts
Roundup (glyphosate)/acre were broadcast over the ex-
periment. On 22 April 1997, the subplot Furadan treat-
ments were imposed by spraying the 1.0 lb ai/acre treat-
ment in a 6-in. band over the row. The Furadan was mixed
with water at a delivery rate of 30 gallon liquid/acre. The
experiment was irrigated within a few hours after applica-
tion of Furadan with 1/3 acre-in. of water to move the
Furadan into the seed zone. On 6 May, 55 lb N/acre was
applied as ammonium nitrate. On 13 May 480 lb 13 (N) -
5 (P2O5) - 29 (K2O) - 1 (Mg) - 2.5 (S)/acre was broadcast
over the experiment. An additional 50 lb N/acre as ammo-
nium nitrate was applied 4 June. Supplemental weed con-
trol was by hooded sprayer, post-direct application of 1.5
pints Gramoxone Extra (paraquat), with non ionic surfac-
tant added at the rate of 1 pint/100 gallon water.
Gramoxone Extra was sprayed in 30 gallon water/acre.
Supplemental gun irrigation water was applied six times at
approximately 1 acre-in. each time during the growing
The two center rows were harvested for fresh ear and
stalk weight on 30 June. Subsamples were taken to deter-
mine dry matter yield. Soil samples for nematode analysis
were collected over each replication and combined at plant-
ing time. Additional samples were collected 18 July from
all plots. Each nematode sample consisted of six cores of
soil (1 in. diameter and 8 in. deep) collected in a system-
atic pattern and then combined into a plastic bag for trans-
port. In the laboratory, a 100-cm3 soil subsample was re-
moved for nematode extraction using a modified sieving
and centrifugation procedure (Jenkins, 1964). Extracted
nematodes were identified and counted under an inverted
microscope. All data were analyzed by an analysis of vari-
ance for a split-plot design, followed by mean separation
by F test or Duncan’s multiple-range test as appropriate.
All sweet corn hybrids responded to application of
Furadan (Table 1). Averaged over all hybrids, fresh ear
weight was 25% greater from the application of Furadan
compared to the control. This same statistic for fresh stalk
weight was a 35% yield increase from application of
Furadan. Fresh ear yield appeared to be greatest for XPH
3084, which was equal to VNE 2 Endeavor. Lowest yields
were obtained by XPH 3105. Average fresh ear yield for
VNE 2 was almost 40% greater than that of XPH 3105,
and with the application of Furadan the difference was
even greater (almost 45%) (Table 1).
In contrast to what one might expect, Furadan did not
reduce nematode numbers as measured 18 July. In fact,
root-knot nematode numbers were over 90% greater in
plots receiving Furadan compared to the control plots
(Table 2). However, of the two highest fresh ear yielding
hybrids, VNE 2 Endeavor, had significantly lower root-
knot nematode counts compared to XPH 3084.
Our data show that sweet corn hybrid selection is criti-
cal if yield is a major factor under consideration (Table
1). With yield increases as much as or more than 35%
from the application of Furadan, it is obvious that this is
one management input that requires consideration by grow-
ers, under conditions of this experiment. These sweet corn
yield responses to application of Furadan are similar to
those found for field corn hybrids (Gallaher and Baldwin,
1985; Gallaher, 1983, 1986a,b; Espaillat and Gallaher,
1989). No information was available regarding type of
pesticide used in the breeding and development of the
sweet corn hybrids used in this study. It is also evident
that Furadan impacted insects or other pests in these sweet
corn hybrids other than the four nematodes measured in
this investigation. It appears that application of Furadan
resulted in an environment that stimulated better plant
growth, which in turn resulted in the healthier plants being
able to tolerate larger populations of root-knot nematodes.
This has been observed and reported for other crops and
cropping systems (McSorley and Gallaher, 1997).
The authors thank Howard Palmer, Walter Davis, John
Frederick and Jacqueline Greenwood for technical assis-
Espaillat, J.R., and R.N. Gallaher. 1989. Corn yield response to tillage,
hybrids, and insecticides. pp. 33-36. In: I.D. Teare, E. Brown
and C.A. Trimble (eds.). Proceedings, 1989 Southern
Conservation Tillage Conference. Special Bulletin 89-1. Inst.
Food and Agr. Sci., University of Florida, North Florida Res. and
Educ. Center, Quincy, Florida.
Gallaher, R.N. 1983. No-tillage corn and sunflower yield response
from Furadan and Counter Pesticides in Alachua County, Florida
in 1982. Agronomy Research Report AY-83-05. Agronomy
Dept., IFAS, Univ. of Florida, Gainesville, Florida.
Gallaher, R.N. 1986a. No-tillage corn response to pesticides, hybrids,
and cropping systems at the Green Acres Agronomy Farm in
1985. Agronomy Research Report AY-86-05. Agronomy Dept.,
IFAS, Univ. of Florida, Gainesville, Florida.
Gallaher, R.N. 1986b. Corn grain yield response to pesticides in
conventional and no-tillage management. Agronomy Research
Report AY-86-09. Agronomy Dept., IFAS, Univ. of Florida,
Gainesville, Florida.
Gallaher, R.N., and J.A. Baldwin. 1985. No-tillage corn results
affected by hybrids and pesticides in 1984 at the Green Acres
Agronomy Farm. Agronomy Research Report AY-85-08.
Agronomy Dept., IFAS, Univ. of Florida, Gainesville, Florida.
Jenkins, W.R. 1964. A rapid centrifugal-flotation technique for
separating nematodes from soil. Plant Dis. Reptr. 48:692.
McSorley, R., and R.N. Gallaher. 1997. Methods for managing
nematodes in sustainable agriculture. pp. 75-79. In R.N. Gallaher
and R. McSorley (eds.). Proceedings 20th Annual Southern
Conservation Tillage Conference for Sustainable Agriculture.
Special Series SS-AGR-60, the Coop. Extension Service, Inst. of
Food and Agr. Sci., Univ. of Florida, Gainesville, Florida.
Norton, D.C., J. Tollefson, P. Hinz and S.H. Thomas. 1978. Corn yield
increases relative to nonfumigant chemical control of
nematodes. J. Nematol. 10:160-166.
Table 1. No-till sweet corn yield for five hybrids at two rates of
carbofuran (Furadan 4F).
Carbofuran rate
Hybrid 1 lb ai1 0 lb ai Average
----- Fresh ear weight, ton/acre -----
XPH 3084 5.37 4.33 4.85 a
VNT 5 Forever 4.38 3.86 4.12 b
VNE 2 Endeavor 5.01 3.70 4.36 ab
VNT 5 Punchline 3.57 2.77 3.17 c
XPH 3105 3.47 2.85 3.16 c
Average 4.36 3.50 *
---Fresh stalk weight, ton/acre---
XPH 3084 8.02 7.23 7.63 a
VNT 5 Forever 7.33 5.06 6.20 b
VNE 2 Endeavor 6.56 4.52 5.54 b
VNT 5 Punchline 6.18 4.04 5.11 b
XPH 3105 3.34 2.48 2.91 c
Average 6.29 4.67 *
-------Dry ear weight, ton/acre-------
XPH 3084 0.93 c 0.81 a NS 0.87
VNT 5 Forever 1.32 b 0.81 a * 1.07
VNE 2 Endeavor 1.61 a 0.92 a * 1.27
VNT 5 Punchline 1.14 bc 0.74 a * 0.94
XPH 3105 1.04 c 0.68 a * 0.86
Average 1.21 0.80
------Dry stalk weight, ton/acre------
XPH 3084 1.86 1.87 1.86 a
VNT 5 Forever 1.89 1.26 1.58 ab
VNE 2 Endeavor 1.64 1.32 1.48 b
VNT 5 Punchline 1.79 1.11 1.45 b
XPH 3105 1.00 0.69 0.84 c
Average 1.64 1.25 *
Data are averages of five replications. Main effect averages in columns
(a,b) not followed by the same letter are different (
= 0.05), according
to Duncan’s multiple-range test. Sub-effect carbofuran with * or NS for
differences at
= 0.05 or not different at
= 0.05, respectively, according
to F test, except for the interaction for dry ear weight, in which case LSD
was used (LSD = 0.23).
1Carbofuran was formulated as Furadan 4F.
Table 2. Effect of carbofuran (Furadan 4F) treatment and
sweet corn hybrid on population levels of plant-parasitic
Nematodes per 100 cm3 soil
18 July
Hybrid 1 April1 + carbofuran2 - carbofuran Average
Ring nematodes,
XPH 3084 123 138 130 a
VNT 5 Forever 154 182 168 a
VNE 2 Endeavor 170 195 183 a
VNT 5 Punchline 122 100 111 a
XPH 3105 214 192 203 a
Average 128 157 161 NS
Root-knot nematodes,
Meloidogyne incognita
XPH 3084 322 130 226 a
VNT 5 Forever 250 123 186 ab
VNE 2 Endeavor 95 39 67 b
VNT 5 Punchline 55 61 58 b
XPH 3105 59 51 55 b
Average 14 156 81 *
Stubby-root nematodes,
Paratrichodorus minor
XPH 3084 5 2 3 a
VNT 5 Forever 4 7 5 a
VNE 2 Endeavor 9 5 7 a
VNT 5 Punchline 2 6 4 a
XPH 3105 5 2 3 a
Average 9 5 5 NS
Lesion nematodes,
XPH 3084 35 36 35 a
VNT 5 Forever 43 39 41 a
VNE 2 Endeavor 22 49 36 a
VNT 5 Punchline 53 57 55 a
XPH 3105 26 37 31 a
Average 10 5 5 NS
Data are means of five replications. Main effect averages in columns
(a,b) not followed by the same letter are different (
= 0.10), according
to Duncan’s multiple-range test. Sub-effect carbofuran with * or NS for
differences at
= 0.10 or not different at
= 0.10, respectively, according
to F test. No interactions were significant at
= 0.10.
1Data from 21 April pooled across all treatments; average of five
2Carbofuran (Furadan 4F) treatments: + = 1.0 lb ai/acre; - = untreated
R.N. Gallaher, J.D. Greenwood and R. McSorley1
The amount of municipal solid waste produced an-
nually in Florida grew to approximately 50 mil-
lion tons in 1992. This represents over 7.9 lb/
resident/day and is twice the national average of about 4
lb/person/day (Smith, 1994). Biodegradable organic waste
that could be composted comprises almost 60% of the
total municipal solid waste. Compostable organic matter
in municipal solid waste includes such things as yard trim-
mings, paper, fast foods, animal manure, crop residues
and food processing residuals. Yard waste trimmings make
up 7.4 million tons annually in Florida (Smith, 1994).
Should all yard waste trimmings be composted, about 4
million tons of compost could be produced annually. In
the U.S., federal law prohibited the use of unlined land-
fills by 1994 (Kidder, 1993). Florida law restricts the
disposal of organic yard waste in lined landfills. These
laws have encouraged a large industry to develop in Florida,
whose objective is to produce wood chip mulch and com-
post from yard waste, often called yard waste compost
(YWC). These products should be environmentally safe to
apply to farmland and result in potential benefits not only
to the farmer but also to society as a whole. For example,
YWC can be applied in large quantities to farmland to
help improve soil properties and crop yield (Gallaher and
McSorley, 1994, 1995; Giordano et al., 1975; Kluchinski
et al., 1993; Mays and Giordano, 1989; Mays et al., 1973;
Wolley, 1995).
Nitrogen is the single-most-important fertilizer input
and is required in the largest quantities for crop produc-
tion (Olson and Sander, 1988). A sweet corn crop has a
sufficient level of N if the concentration in the diagnostic
ear leaf at full silking and tasseling is between 2.5 and
3.0% (Jones et al., 1991). Normal N fertilizer recom-
mendations may differ significantly for crops grown on
soils having received large quantities of YWC or other
biodegradable organic waste product. Legumes are known
to contain significant quantities of N and other fertilizer
elements and can serve as sources of organic fertilizer
(Wade et al., 1997; Wieland et al., 1997; Xiao et al., 1998).
Soil K and Mg increased and diagnostic leaf N and P
concentrations increased as cowpea (Vigna unguiculata
L.) pod yield increased with increasing rates of lupin hay
(Wieland et al., 1997). Studies showed that application of
2 to 3 tons of lupin (Lupinus angustifolius L.) hay/acre
would maximize cowpea yield (Wieland et al., 1997; Xiao
et al., 1998). In another study, bushbean pod yield reached
maximum at 2 tons/acre crimson clover (Trifolium
incarnatum L.) hay (Wade et al., 1997). The objective of
this study was to investigate the changes in soil properties
and impact on sweet corn (Zea mays L.) yield from use of
YWC at five rates of lupin hay as an organic source of N
and other nutrients.
This research was conducted the fifth year (1997) fol-
lowing application of 120 ton YWC/acre each year for the
previous four years (Table 1). A winter cover crop of ‘Tift
Blue’ lupin was mowed closely just prior to planting sweet
corn. ‘Silver Queen’ sweet corn was planted at 28,000
plants/acre in four-row plots, 30 in. wide and 12 ft long
using a Brown-Harden in-row subsoil no-till (strip-till)
planter. Seeders on the planter were John Deere Flexie
71’s. The three main-plot treatments were residual YWC
cumulative treatments (480 ton YWC/acre no-till; 480
ton YWC/acre conventional tillage; conventional tillage
control) from the past four years. No additional YWC was
applied in 1997 prior to planting this experiment. Sub-
plots were five rates of lupin hay (0, 2, 4, 6, 8 ton/acre) as
a source of organic fertilizer, either incorporated just prior
to planting or side-dressed as a mulch immediately after
planting. All treatment combinations were replicated four
times. The crop was hand hoed for weed control as needed.
Approximately 1 acre-in. of irrigation was applied six
times. No chemical management inputs were made. Ear
leaf samples were collected at early silking and analyzed
for N concentrations (Gallaher et al., 1975). Soil samples
were collected from the top 8 in. in February prior
toplanting corn and in August following corn harvest. Soil
samples were analyzed for extractable nutrients, pH, or-
ganic matter and cation exchange capacity. Soils were fur-
ther analyzed for plant-parasitic nematodes using appro-
priate procedures (McSorley and Gallaher, 1997; and
Jenkins, 1964). All data were analyzed by analysis of vari-
ance for a split-plot design, followed by mean separation
by Duncan’s multiple-range test.
1University of Florida, Institute of Food and Agricultural Sciences, Agronomy
Department (Gallaher and Greenwood) and Entomology and Nematology
Department (McSorley), Gainesville, Florida.
The residual impact of application of large quantities
of YWC, including large quantities of plant nutrients and
organic matter (Table 1), resulted in significant improve-
ment in soil quality for the beginning of this investigation
(Table 2). Little differences existed between the two YWC
treatments, but both were several hundred percent greater
in most properties than the control. This condition per-
sisted throughout the duration of the experiment, as evi-
denced by the summer soil test that followed (Table 2). It
is obvious that the previously applied YWC should have a
significant impact on crops growing under these condi-
Yield of fresh and dry sweet corn ears shows that YWC
was effective in increasing yield as evidenced by the in-
tercept for the three YWC treatments (Fig. 1). Yield was
highest for the YWC treatment when corn was grown un-
der no-tillage management, intermediate for YWC con-
ventional tillage and least for the control. These differ-
ences among the three YWC treatments were consistent
across all five lupin hay rates (Fig. 1). Data show that
maximum fresh ear yield was achieved at about 2 ton lupin
hay/acre for YWC no-till treatment, about 4 ton lupin hay/
acre for YWC conventional tillage treatment and about 6
ton lupin hay/acre for the conventional tillage control treat-
ment. One possible explanation for the higher yields for
YWC no-till treatment is the likely conservation of soil
water from the lack of soil disturbance as well as lupin
hay mulch benefits. Lupin hay on the soil surface would
also result in slower release of plant nutrients in the hay,
and thus reduce the potential for excessive leaching of
nutrients out of the root zone during heavy rainfall events,
as compared to incorporation.
Nitrogen concentration in the ear leaf (Table 3) was
highly correlated with ear yield (Fig. 1) and was directly
caused by increasing rates of lupin hay (Table 3). Suffi-
ciency levels (Jones et al., 1991) for N in the ear leaf
were achieved for both YWC treatments at 2 ton YWC/
acre but required 6 ton YWC/acre for the control. This
evidence, along with yield data, further supports the fact
that residual effects of YWC not only improve soil qual-
ity but also provide an environment for increased crop
yield and leaf quality. It also indicates that lupin hay is a
possible source of organic fertilizer in all of the YWC
Nematodes were not affected either by YWC treat-
ments or by the application of lupin hay (Table 4). Signifi-
cant quantities of ring and root-knot nematodes were
present, and both increased in numbers by the end of the
crop season. Therefore, yield differences in this study
were not the result of the nematodes measured.
The authors thank Jim Chichester, Howard Palmer,
Walter Davis and John Frederick for technical support and
Mr. Tom Fulmer of Enviro-Comp Services, Inc., Jackson-
ville, Florida, for providing the yard waste compost used
in this study.
Gallaher, R.N., and R. McSorley. 1994. Soil water conservation from
management of yard waste compost in a farmer’s corn field.
Agronomy Res. Rept. AY-94-02. Agronomy Dept., Inst. Food
& Agr. Sci., Univ. of Florida, Gainesville, Florida.
Gallaher, R.N., and R. McSorley. 1995. Conventional and no-tillage
vegetables using yard waste compost. Agronomy Res. Rept.
AY-95-01. Agronomy Dept., Inst. Food & Agr. Sci., Univ. of
Florida, Gainesville, Florida.
Gallaher, R.N., C.O. Weldon and J.G. Futral. 1975. An aluminum
block digester for plant and soil analysis. Soil Sci. Soc. Amer.
Proc. 39:803-806.
Giordano, P.M., J.J. Mortvedt and D.A. Mays. 1975. Effect of
municipal wastes on crop yields and uptake of heavy metals.
Journal of Environmental Quality 4:394-399.
Jenkins, W.R. 1964. A rapid centrifugal-flotation technique for
separating nematodes from soil. Plant Dis. Reptr. 48:692.
Jones, J.B., Jr., H.A. Mills and B. Wolf. 1991. Plant Analysis
Handbook. Micro-Macro Publishing, Inc., Athens, Georgia.
Kidder, G. 1993. Applying non-hazardous wastes to land: I.
Opportunities and problems. SS-SOS-43, Notes in Soil and
Water Science, Univ. of Fla., Inst. Food & Agr. Sci., Coop. Extn.
Serv., Gainesville, Florida.
Kluchinski, D., J.R. Heckman, J. Mahar, D.A. Derr and F. Kelly.
1993. Leaf mulching: Using municipal leaf waste as a soil
amendment. Special Report by New Jersey Agric. Exp. Station,
Rutgers Univ., Trenton, New Jersey.
Mays, D.A., and P.M. Giordano. 1989. Land spreading municipal
waste compost. Biocycle 30:37-39.
Mays, D.A., G.L. Terman and J.C. Duggan. 1973. Municipal
compost: Effects on crop yields and soil properties. Journal of
Environmental Quality 2:89-92.
McSorley, R., and R.N. Gallaher. 1997. Methods for managing
nematodes in sustainable agriculture. pp. 75-79. In: R.N.
Gallaher and R. McSorley (eds.). Proc. 20th Annual Southern
Conservation Tillage Conference for Sustainable Agriculture.
Special Series SS-AGR-60, University of Florida, Inst. Food &
Agr. Sci., Coop. Extn. Serv., Gainesville, Florida.
Olson, R.A., and D.H. Sander. 1988. Corn production. pp. 639-686.
In: G.F. Sprague and J.W. Dudley (eds.). Corn and corn
improvement. Number 18 in the Agronomy Series. American
Society of Agronomy, Inc., Crop Science Society of America,
Inc. and Soil Science Society of America, Inc., Madison,
Smith, W.H. 1994. Recycling composted organic materials in Florida.
Florida Coop. Extn. Serv., Inst. Food & Agr. Sci., Univ. of
Florida, Gainesville, Florida. BP-2, July 1994.
Wade, B.L., S.J. Rymph and R.N. Gallaher. 1997. Assessment of soil-
incorporated crimson clover hay as an organic fertilizer source in
the production of bush bean. pp. 108-116. In: R.N. Gallaher and
R. McSorley (eds.). Proc. 20th Annual Southern Conservation
Tillage Conference for Sustainable Agriculture. Special Series Table 1. Analysis of yard waste compost used on the Green
SS-AGR-60, University of Florida, Inst. Food & Agr. Sci., Coop. Acres Agronomy Field Laboratory research plots.
Extn. Serv., Gainesville, Florida. Year
Wieland, C.E., J.A. Widmann and R.N. Gallaher. 1997. Lupin hay as Analysis 1993 1994 1995 1996
an organic fertilizer for production of ‘White Acre’ cowpea. pp. DM %1 45.1 49.8 50.7 57.7
100-107. In: R.N. Gallaher and R. McSorley (eds.). Proc. 20th OM % 48.2 59.2 42.2 52.2
Annual Southern Conservation Tillage Conference for C % 33.5 31.3 33.5 32.0
Sustainable Agriculture. Special Series SS-AGR-60, University N % 0.81 0.91 0.98 0.63
C:N ratio 41.7 34.4 36.4 50.8
Wolley, J.S., Jr. 1995. The switch from conventional to sustainable. pH ground
Ca % 6.3
3.43 7.1
3.41 7.0
1.14 6.2
Resource, April 1995:7-9. Mg % 0.18 0.19 0.07 0.17
Xiao, Y., M.W. Edenfield, E. Jo and R.N. Gallaher. 1998. Production K % 0.22 0.29 0.14 0.31
and leaf nutritional response of ‘White Acre’ cowpea (Vigna P % 0.17 0.18 0.08 0.15
unguiculata) to management strategies of perennial peanut Cu ppm 23.0 18.0 18.0 22.0
of Florida, Inst. Food & Agr. Sci., Coop. Extn. Serv., Gainesville, pH chopped 6.2 7.5 --- 6.5
(Arachis glabrata) hay as an organic fertilizer. Agronomy Res. Fe ppm 1953.0 1825.0 2608.0 2615.0
Rept. AY-98-01. Agronomy Dept., Inst. Food & Agr. Sci., Univ. Mn ppm 180.0 188.0 75.0 97.0
of Florida, Gainesville, Florida. Zn ppm 102.0 118.0 138.0 148.0
1DM % = dry matter; OM % = organic matter in DM; chopped = compost
samples were chopped into coarse particles using a grinder; ground =
sub-samples of the chopped samples were ground with a Wiley mill to
pass a 2-mm stainless steel screen. Values are the average of four
replications. The source of the compost was Wood Resource Recovery,
Gainesville, Florida, from 1993 to 1995 and Enviro-Comp Services Inc.,
Jacksonville, FL in 1996.
Table 2. Mehlich I extractable elements, Kjeldahl N and other soil analyses after yearly application
of 120 ton yard waste compost/acre/year from 1993 to 1996.
Cumulative Yard Waste Compost-YWC (120 ton/acre/year)
No-till Conv-till Conv-till
Analysis Unit LSD CV 480 ton/acre 480 ton/acre 0 ton/acre
Winter 1997, no YWC added in 1997, test prior to planting sweet corn
N ppm 448 21.7% 1613 1530 442
P ppm 12 6.3% 140 132 67
K ppm 15 20.0% 52 49 25
Na ppm 4.3 15.2% 20 19 10
Ca ppm 566 21.4% 2163 2042 374
Mg ppm 36 17.7% 158 151 46
Cu ppm 0.14 19.6% 0.30 0.33 0.61
Fe ppm 1.03 14.3% 3.8 4.5 4.1
Mn ppm 1.74 13.0% 10.5 9.9 2.8
Zn ppm 2.57 13.6% 14.6 14.1 4.0
pH 0.15 1.3% 6.9 6.8 6.6
BpH NS 0.3% 7.88 7.86 7.86
OM % 1.21 21.4% 4.38 4.18 1.31
CEC meq/100g 3.18 18.6% 13.35 12.80 3.5
Summer 1997, test following sweet corn harvest
N ppm 440 23.2% 1063 1123 428
P ppm 22 11.6% 122 126 84
K ppm NS 41.6% 40 32 37
Na ppm 9.6 15.1% 44 37 30
Ca ppm 676 30.2% 1709 1834 336
Mg ppm 45 28.7% 115 121 39
Cu ppm 0.20 31.2% 0.30 0.33 0.52
Fe ppm 2.47 16.1% 7.3 8.5 10.8
Mn ppm 3.29 22.1% 10.1 11.2 4.5
Zn ppm 4.05 23.6% 12.2 13.4 4.1
pH 0.20 2.2% 6.8 6.7 6.2
BpH NS 0.3% 7.83 7.82 7.79
OM % 0.64 21.5% 4.14 3.82 1.26
CEC meq/100g 3.96 25.5% 11.15 11.86 3.89
Table 3. Nitrogen concentration in ear leaf of ‘Silver Queen’
sweet corn from yard waste compost and lupin treatments.
Yard Waste Compost Treatments
Lupin Hay No-Till Conv-Till Control
tons/acre -------------------------- % N -----------------------------
0 2.40 L 2.49 L 1.87 L
2 2.51 S 2.62 S 2.43 L
4 2.71 S 2.73 S 2.49 L
6 2.82 S 2.83 S 2.74 S
8 2.74 S 2.70 S 2.74 S
= 0.05) = 0.28; CV = 7.4%; No-till and Conv-till treatments received
a cumulative total of 480 tons yard waste compost/acre in 120 ton/acre/
year increments from 1993 to 1996. No compost was applied in 1997.
L = low and S = sufficient levels of N in ear leaf according to Jones et
al., 1991.
Table 4. Effect of yard-waste compost on nematode
population levels in plots of ‘Silver Queen’ sweet corn, 1997.
Sampling Date
Compost Treatment 6 March 28 July
Nematodes per 100 cm3 soil
Ring (
Mulch, No-till 66 143
Incorporated, Conventional-till 66 399
Control, Conventional-till 105 328
Root-knot (
Meloidogyne incognita
Mulch, No-till 24 222
Incorporated, Conventional-till 10 172
Control, Conventional-till 10 172
(Paratrichodorus minor)
Mulch, No-till 1 1
Incorporated, Conventional-till 0 4
Control, Conventional-till 2 3
Lesion (
Mulch, No-till 12 11
Incorporated, Conventional-till 20 24
Control, Conventional-till 31 25
Data are means of four replications. No significant treatment effects at
< 0.10.
Compost applied as mulch or incorporated, both treatments at 480 ton/
Fig. 1. Silver Queen sweet corn, fresh and dry ear weights;
YWC = yard waste compost; No-till = strip-till; C-till =
conventional tillage; the YWC treatmenst were from
residual applications from the previous four years.
E.C. Gordon, T.C. Keisling, D.M. Wallace, L.R. Oliver and C.R. Dillon1
Experiments were conducted at the Northeast Re-
search and Extension Center (NEREC) at Keiser,
Arkansas, in 1995 and 1996 to determine the influ-
ence of tillage system, planting date and cultivar selection
on soil water storage, soybean (Glycine max, L. Merr.)
yield and economics. The soil series was Sharkey silty
clay. ‘Williams 82’, ‘Manokin’ and ‘RA 452’ soybean cul-
tivars were planted in mid-April, and RA 452, ‘Pioneer
9592’ and ‘Pioneer 9641’ were planted in mid-May, mid-
June and mid-July. The cultivars were stripped in three
production systems consisting of no-till, fallow and con-
ventional. Soil water levels were monitored gravimetri-
cally in each tillage system weekly to a depth of 60 cm.
The Sharkey silty clay maintained high soil water storage
of 8 to 10 cm in the 0- to 60-cm depth. Sharkey silty clay
was able to maintain high soil water for April- and May-
planted soybean. The adequate soil water resulted in high
yields for April- and May-planted soybean with the early
maturity-group cultivars, Williams 82 and RA 452. De-
layed planting dates conserved soil water and resulted in
the highest soybean yields in June- and July-planted soy-
bean with Pioneer 9592 and Pioneer 9641. The June no-
till production system had the highest costs because of
high herbicide usage. The highest net returns corresponded
to the highest soybean yields. Overall, under a conven-
tional production system on a Sharkey silty clay, the most
profit was obtained when an early maturity group soybean
was planted in April or May.
Dryland soybean production encompasses approximately
65% of soybean (Glycine max, L. Merr.) grown in Arkan-
sas. The low profitability of soybean relative to some other
enterprises has resulted in increased interest in minimum
input production systems. The common occurrence of a
drought in the mid-South from mid-July to mid-Septem-
ber has contributed to low and stagnant yields in dryland
soybean (Bowers, 1995; Heatherly, 1996). Commonly
planted Maturity Group V and VI cultivars are in the criti-
cal reproductive stages during the late-season drought, and
their yield potential can be greatly reduced by these
droughts (Miller, 1994). Dryland producers subjected to
1Res. Assoc. and Prof. of Agron. located at NEREC, Keiser, Arkansas. Former
Grad. Student, Prof. of Agron., Assoc. Prof. Of Agric. Econ. located at
Fayetteville, Arkansas, respectively.
the possibility of drought during a growing season require
a production system to avoid or tolerate the effect of a
drought. Manipulation of practices such as tillage system,
planting date and cultivar selection could potentially in-
crease soybean yield under dryland conditions.
Tillage Practices
Typical soybean production in the mid-South includes
some type of mechanical tillage for seedbed preparation
(Bowers, 1995). The general purpose of conventional till-
age is to control weeds and create a favorable environ-
ment for seed emergence and plant growth. Conventional
tillage provides a tilled soil layer of 15 to 25 cm deep.
No-till is a cropping system in which the soil is left un-
disturbed prior to planting, and weed control is accom-
plished by herbicides. No-till systems are associated with
conservation tillage, which is defined as a tillage and plant-
ing system that maintains at least 30% of the soil surface
covered by residue at the time of crop emergence (Dick
et al., 1989; Parsch et al., 1993).
Different management practices result in varying costs
of production. Webber et al. (1987) noted that no-till pro-
duction systems reduce soil erosion, decrease overall fuel
consumption and equipment costs and conserve soil mois-
ture. Although no-till generally saves fuel, labor and ma-
chinery costs, total costs may be higher due to increased
herbicide expenditures as compared to conventional sys-
tems (Letey, 1984).
Planting Date
Soybean production in the mid-South has been prima-
rily limited to Maturity Group V and VI cultivars, which
are planted in May and June. Yield reductions due to
drought stress occur in these cultivars quite often, be-
cause they are blooming, setting pods and beginning seed
fill during July and August when there is a high probability
of soil moisture deficit. Changing the planting date to an
earlier or later time would shift the time when soybean
plants bloom, set seed and mature, thus creating the possi-
bility that moisture stress could be avoided during these
critical periods. In the mid-South, higher rainfall amounts
occur in the spring and fall with the greatest spring rain-
fall occurring from April to early June. This corresponds
with early bloom and pod set in April-planted, early matu-
rity, indeterminate and determinate soybean cultivars (Bow-
ers, 1995; Miller, 1994). The early maturity group culti-
vars experience cooler temperatures and lower evapora-
tive demand, which reduces overall water demand. The
ability of early maturity group cultivars to bloom and set
pods under milder temperatures with adequate moisture
increases the chance of profitable yields (Board and Hall,
1984; Heatherly, 1996; Miller, 1994).
Planting early-maturing cultivars, however, has disad-
vantages. Cool temperatures can delay emergence and re-
tard growth rate. Planting dates may also be delayed due
to spring rains, and a reduction in seed quality can occur
(Unger and Cassel, 1991). Weed control problems at leaf
drop may be associated with early Maturity Group III and
IV cultivars (Dombek et al., 1995; More, 1994, Parsch et
al., 1993). This can create harvesting problems and neces-
sitate the extra cost of a desiccant application.
Research that has been conducted on late plantings of
soybean has provided lower yield results for July planting
dates as compared to May and June planting dates (Hancock,
1994; Moore, 1994). Some research at many locations
suggests that day length, not water stress, is responsible
for the declining yield after mid-June, since the yield re-
duction could not be eliminated with irrigation (Beuerlein,
1988; Board and Hall, 1984; Reeves and Tyler, 1996).
Board and Hall (1984) have shown that a major reason for
yield losses at nonoptimal planting dates is inadequate
vegetative growth due to premature flowering, but yield
losses due to late planting dates vary by year.
Indeterminate growth characteristics are being utilized
more in southern cultivar selection. The main difference
in growth habit between the determinate and indetermi-
nate types is that indeterminate cultivars continue main
stem elongation several weeks after the plants begin to
flower; whereas determinate cultivars halt elongation of
the main stem at the onset of flowering (Beuerlein, 1988).
Indeterminate cultivars can cease growth temporarily and
then restart when stress is removed. These growth charac-
teristics may be important factors for soybean grown in
the mid-South due to prolonged drought conditions.
Soil Moisture
Tillage systems influence soil water content through
infiltration and runoff, evaporation and precipitation stor-
age. Evaporation from a soil is affected by the residues
left on the soil surface and by the soil properties. Tillage
alters infiltration and runoff through surface residue, bulk
density and soil crusting.
Soil crusts may develop on no-till and conventionally
tilled soils, reducing water infiltration and increasing run-
off. Water infiltration and runoff are also influenced by
surface residue and bulk density. Soils with high residue
prevent the formation of soil crusts. If soil residue is
adequate, surface infiltration will be enhanced. Soils with
low residue levels require tillage for enhanced infiltration
(Unger and Cassel, 1991).
Conventional tillage may promote degradation of the
soil physical condition by reducing the soil pore volume
and water storage area (Letey, 1984). Tillage increases
the susceptibility of the soil to compaction by traffic or
natural consolidation. Plants growing in soils with tillage
pans may undergo severe moisture stress after 5 to 8 days
without rainfall (Reeves and Tyler, 1996).
Conservation tillage results in greater compaction of
the top 10 cm of soil as compared to conventional tillage.
However, this compaction can prevent more severe com-
paction at greater depths (Reeves and Tyler, 1996). Soils
with less-available moisture favor high yields in early-
maturity group cultivars whereas deep soils favor high
yields in late maturity group cultivars (Miller, 1994).
The objective of this research was to evaluate cultural
practices, including tillage practice, planting date and cul-
tivar selection, for potential to increase soybean yield and
profitability under dryland conditions.
Field experiments were conducted in 1995 and 1996 at
the Northeast Research and Extension Center at Keiser,
Arkansas, on a Sharkey silty clay soil series. The experi-
mental design was a split-split strip plot with four replica-
tions. The individual plot size was 3 m wide by 7 m long
with 9-m alleys. The main plot was four planting dates:
mid-April, mid-May, mid-June and mid-July. Subplots were
tillage levels: no-till, fallow and conventional. Three soy-
bean cultivars were stripped within each tillage level. The
tillage subplots had a 3-m fallow border between tillage
systems. The plots were not irrigated. Weather data were
collected at the location, and all production inputs were
recorded by planting date and production practice.
Tillage levels were based on practices that potentially
conserve soil moisture. No-till plots were not disturbed
from the fall prior to experiment establishment until the
conclusion of the experiment. The fallow treatments were
tilled 3 to 5 cm deep with a roto-tiller following each
rainfall event prior to planting. Conventionally tilled plots
were tilled 10 to 15 cm deep in the fall and prior to
soybean planting or when vegetation reached a height of
15 to 24 cm.
Herbicide programs were designed for complete weed
control (Table 1). Two weeks prior to planting, the no-till
system received a burndown application of glyphosate
(Roundup Ultra®) to desiccate winter weeds and emerging
summer annuals. The no-till and fallow systems then re-
ceived applications of metolachlor (Dual II®)+ a premix
of metribuzin and chlorimuron (Canopy®) applied preemer-
gence. A preplant incorporated application of trifluralin
(Treflan®) + metribuzin and chlorimuron (Canopy®) was
applied to the conventional system. All tillage systems
received fomesafen (Reflex®) as a post-emergence over-
the-top application as needed for weed control during the
growing season. Dates of post-emergence herbicide ap-
plications varied and are presented in Table 2.
Cultivars were selected from the Arkansas Variety Se-
lection Program (Dombek et al., 1995) and varied with
planting date (Table 3). Cultivars in Maturity Groups III
and IV were selected for the mid-April planting date, and
cultivars in Maturity Groups IV, V and VI were used in the
mid-May, mid-June and mid-July planting dates. Both in-
determinate and determinate cultivars were used in the
cultivar selection.
Soybean seeds were planted flat in 18-cm row spacing
with a 3-m-wide John Deere no-till drill. Seeding rate was
9 to 12 seeds/m of row. Plots were harvested with a plot
combine at maturity.
Soil moisture in the tillage production systems was
measured gravimetrically at planting and every week dur-
ing the growing season, except after rainfall when soils
were saturated. Soil samples were taken at random to a
depth of 8 cm from each tillage method plot at planting
and after planting in 1995. In 1996, soil samples were
taken to a depth of 60 cm. Soil sampling was discontinued
when the earliest maturing cultivar in the planting date
reached the R6 growth stage.
Economic analysis of the experiment was conducted
using the Mississippi State University Budget Generator
computer program. All economic inputs were recorded
and entered. Variable and total costs were generated along
with net returns. The average price of soybean used in the
economic analysis to calculate net returns was $5.92/bu.
All data were subjected to analysis of variance using
the GLM (General Linear Model) procedure of SAS. Means
were separated using Fisher’s Protected LSD (0.05).
Soybean yields, economic costs and net returns could
be pooled over years. Environmental conditions varied little
between years. Rainfall levels were higher in 1996 but did
not significantly affect soil water storage or soybean yield.
Tillage level had few significant influences on soil wa-
ter storage and soybean yield (data not shown). The tillage
levels implemented were expected to alter soil water
evaporation rates and soil water storage (Mwendera and
Feyen, 1994). However, the shallow tillage operations
could not be conducted immediately after rainfall due to
travel and labor restrictions, and some evaporation oc-
curred before the implementation of the fallow tillage
system. Consequently, soil water samples were taken af-
ter evaporation losses in each production system.
Soil samples for soil water storage determination were
taken randomly by planting date each year. As a result,
years could not be combined by sampling date and will be
discussed separately. Soil samples for soil water storage
determination were taken from only a 0- to 8-cm depth in
1995, and there was no influence on soil water storage or
soybean yields due to the shallow sampling depth. In 1996,
the soil was sampled to a depth of 60 cm with a new
sampling technique utilizing lubricants, and only these data
will be discussed.
Planting date significantly affected soil water storage
and soybean yield and will be discussed by specific plant-
ing date. Also, cultivar selection significantly affected soy-
bean yields at the varying planting dates. Therefore, indi-
vidual cultivar yields will be discussed within a planting
date. Soil water sampling was taken at random across the
three cultivar strips for each tillage level. Therefore, cul-
tivars and their effects on soil water storage and econom-
ics could not be evaluated.
Soil Water Storage
In 1996, soil water storage was similar among the April,
May and June planting dates (Fig. 1). Frequent rainfall
replenished soil water levels until August. However, some
variation in soil water levels was observed in June and in
the duration of drought during each planting date.
The April planting date had the lowest soil water stor-
age in mid-June to mid-July (Fig. 1). Since soil water
utilization began in April, the April-planted soybean roots
had removed soil water for the longest duration. Drought
conditions did not occur until August, allowing the April-
planted soybean to reach maturity before severe water
stress. These results coincide with the findings of Bowers
(1995) and Miller (1994).
The May and June planting dates maintained slightly
higher soil water levels in June and July than the plots
planted in April (Fig. 1). The May and June planting dates
conserved soil water in April and May that could be used
in June and July. In August, the May and June planting
dates decreased dramatically in soil water. Drought condi-
tions resulted in the use of all available soil water.
The July planting date maintained the highest soil water
storage in August during the drought conditions (Fig. 1).
The delayed planting date allowed soil water conservation
in April, May, and June in the absence of vegetation. Pre-
vious research (Hancock, 1994) showed that weed-free
areas have higher soil water storage.
Soybean Yields
The April- and May-planted soybean had the highest
yields (Table 4). Soybean yields decreased when the plant-
ing date was delayed due to drought and decreasing photo-
The Maturity Group III cultivar, Williams 82, yielded
the highest of the April-planted cultivars. The early matu-
rity cultivar matures during the highest soil water storage
levels, and its indeterminate growth patterns can increase
vegetative growth, which can increase soybean yield. There-
fore, Maturity Group III cultivars can avoid water stress
and produce high yields (Bowers 1995; Heatherly, 1996;
Miller, 1994). The Maturity Group IV cultivars, RA 452
and Manokin, have a longer growing season, which ex-
tended the reproductive stage into drought conditions for
a longer duration (Fig. 1) and affected yield.
RA 452, a Maturity Group IV cultivar with indetermi-
nate growth, had the lowest yield of the cultivars planted
in the delayed planting dates due to premature flowering
(Table 4). Pioneer 9592, a Maturity Group V cultivar with
determinate growth, had the highest yields and was the
best-suited cultivar for the May and June planting dates.
Pioneer 9641, a Maturity Group VI cultivar with determi-
nate growth characteristics, had the longest growing sea-
son of the cultivars and the lowest yields when planted in
May and June due to dry conditions during its reproduc-
tive period. However, when planted in July, Pioneer 9641
had the highest soybean yields.
Economics Costs
The conventional production system had the lowest
costs (Table 5). Mechanical preplant tillage operations
for weed control in conventional tillage resulted in lower
production costs than equivalent herbicide programs in
The fallow production system costs were slightly higher
than the conventional production system. The fallow-till-
age system had shallow tillage after rainfall events of >2
cm to destroy soil crusts. Shallow tillage was often per-
formed two or three times a month during frequent rain-
fall events. Thus, the high number of tillage operations
increased costs in the fallow production system as com-
pared to the conventional production system.
The no-till production system had the highest costs
(Table 5), because no-till required the application of a
preplant burndown herbicide for adequate weed control.
The preplant burndown herbicide application was more
costly than mechanical tillage, resulting in higher variable
and total costs than the conventional tillage production
The June planting date, regardless of tillage system,
had the highest variable and total costs and July the lowest
of the planting dates under fallow and conventional pro-
duction systems (Table 5). This was due to weed pressure,
which necessitated post-emergence applications for June
planting dates, while July planting dates required only pre-
plant or preemergence herbicide applications (Table 2).
The lowest production variable and total costs in the no-
till production system were in April due to low herbicide
costs (Table 5).
Net Returns
Production systems greatly influenced net returns (Table
6). The no-till system provided the lowest net returns due
to higher herbicide costs. The slight increase in costs of
the fallow production system did not affect net returns,
since the fallow system had slightly higher net returns
than the conventional system for all planting dates except
June. The high cost of the no-till production system re-
sulted in a decrease of approximately $80/ha and $62/ha
in net returns as compared to the fallow and conventional
production systems, respectively.
Average net returns were the highest in April and May
planting dates (Table 6). After May, net returns decreased
sharply, becoming the lowest in July. A relatively low
range occurred in soybean yields between years, and the
planting dates with the highest net returns should be used.
To achieve the lowest risk in soybean production and high-
est average net returns, the planting dates for soybean
should spread out among all the planting dates.
Soil Water Storage
The Sharkey silty clay maintained approximately 8 to
10 cm of soil water to a 60-cm depth. Thus, April- and
May-planted soybeans on the Sharkey silty clay poten-
tially avoided drought stress by maturing before soil water
was depleted in the root zone. Cumulative water removal
of early-planted soybean resulted in low soil water levels
during July and August under drought conditions. Main-
taining a vegetation-free surface conserved soil water,
which could subsequently be used by late-planted (June
and July) soybean. This would be especially important dur-
ing seasons with prolonged drought periods.
Soybean Yields
Soybean yields were influenced by planting date and
cultivar selection. April- and May-planted soybean plots
yielded the highest with the Maturity Group III indetermi-
nate Williams 82 being the best for April planting. The
Pioneer 9592 Maturity Group V determinate cultivar was
best suited for May planting. RA 452, a Maturity Group
IV indeterminate cultivar, also had high yields when planted
in May. Soybean yields typically declined in June and July
planting dates relative to April and May plantings. Pioneer
9592 should be planted in May. June and July planting
dates should be avoided.
No-till production systems were always more expen-
sive than the fallow or conventional production systems.
Tillage operations cost less than herbicide applications
for weed control. Planting dates influenced costs because
of herbicide requirements with the June planting date hav-
ing the highest cost. High weed pressure in June required
repeated applications of postemergence herbicides and re-
sulted in high herbicide costs. The lowest cost occurred
in the July planting date, which did not have to rely on
post-emergence herbicide applications.
A no-till production system resulted in approximately
a $80/ha and $62/ha loss in net returns as compared to the
fallow and conventional production systems, respectively.
- - -
---- ---- ----
---- ---- ----
The net returns at each planting date followed the same
trend as soybean yields. April- and May-planted soybeans
had the highest yields and highest net returns.
Beuerlein, J.E. 1988. Yield of indeterminate and determinate
semidwarf soybean for several planting dates, row spacings, and
seeding rates. J. Prod. Agric. 1:300-303.
Board, J.E., and W. Hall. 1984. Premature flowering in soybeans and
yield reduction at nonoptimal planting dates as influenced by
temperature and photoperiod. Agron. J. 76:700-704.
Bowers, G.R. 1995. An early soybean production system for drought
avoidance. J. Prod. Agric. 8:112-119.
Dick, W.A., R.J. Rosenberg, E.L. McCoy, W.M. Edwards and F.
Haghiri. 1989. Surface hydrologic response to soils in no-tillage.
Soil Sci. Soc. Am. J. 53:1520-1526.
Dombek, D.G., R.D. Bond and S.B. Cain. 1995. Arkansas soybean
performance tests 1994. Univ. Ark., Variety Testing Publ. 2055.
Hancock, F.G. 1994. Using row spacing and planting date to your
advantage. Proc. South. Soy. Conf. 2:138-140.
Heatherly, L.G. 1996. Performance of MG IV and V soybeans in early
and conventional plantings in the Mid-South. Proc. South. Soy.
Conf. 4:6-10.
Letey, J. 1984. Relationship between soil physical properties and crop
production. Soil Sci. 1:277-294.
Miller, T.D. 1994. Why early soybeans? A summary of the Texas
experience. Proc. South. Soy. Conf. 2:103-105.
Moore, S.H. 1994. Potential for increasing soybean yield at late
planting dates using cultivars with indeterminate stem growth and
delayed flowering. Proc. South. Soy. Conf. 2:192-197.
Mwendera, E.J. and J. Feyen. 1994. Effects of tillage and evaporative
demand on the drying characteristics of a silt loam: An
experimental study. Soil and Tillage Res. 32:61-69.
Parsch, L.D., N.S. Crabtree and L.R. Oliver. 1993. Economics of no-
till and conventional tillage for soybean crop rotations. Proc.
South. Soy. Conf. 2:109-114.
Reeves, D.W., and D.D. Tyler. 1996. Soybean production in the
reduced tillage system: Soil compaction overview. Proc. South.
Soy. Conf. 4:202.
Unger, P.W., and D.K. Cassel. 1991. Tillage implement disturbance
effects on soil properties related to soil and water conservation:
A literature review. Soil and Till. Res. 19:363-382.
Webber, C.L., H.D. Keff and M.R. Gebhardt. 1987. Interrelations of
tillage and weed control for soybean (Glycine max) production.
Weed Sci. 35:830-836.
Table 1. Herbicide programs.
Method of
Trade name Common name application1 Rate
Roundup Ultra2 glyphosate PPBD 1.12
Dual II + Canopy3 metolachlor + PRE 2.8
Treflan + Canopy4 trifluralin + PPI 1.12
Ref lex 3 fomesafen POST 0.42
1Method of application: PPBD = preplant burndown, PPI = preplant
incorporated, PRE = preemergence, POST = postemergence.
2Treatments used only in no-till tillage system.
3Treatments used only in no-till and fallow tillage systems.
4Treatments used only in conventional tillage systems.
Table 2. Postemergence herbicide applications.
Application timing and soybean stage
Planting 1995 1996
date Herbicide Date Stage Date Stage
April Reflex 6/21 V5 6/27 V5
May Reflex 6/21 V3 6/27 V3
June Reflex 7/08 V2 7/14+7/25 V2+V3
July Reflex 7/25 V2
Table 3. Planting date and cultivar selection.
Planting Maturity Growth
date Cultivar group characteristics1
Mid-April Williams 82 III ID
Manokin IV D
Ring Around 452 IV ID
Mid-May, Ring Around 452 IV ID
Mid-June, Pioneer 9592 V D
Mid-July Pioneer 9641 VI D
1ID = indeterminate; D = determinate.
Table 4. Influence of planting date and cultivar
on average soybean yield.
Planting date
Cultivar April May June July
Williams 82 3516
Manokin 3289
RA 452 3245 3301 2425 1193
Pioneer 9592 3559 2624 1565
Pioneer 9641 3173 2386 1753
for comparing among planting dates = 161
for comparing among cultivars = 134
Table 5. Influence of planting date and tillage system on average variable and total economic costs at Keiser (1995 and 1996).
Tillage Variable costs Total costs
system April May June July April May June July
---------------------------------------------------------------------------- ($/ha) -------------------------------------------------------------------------------
No-till 260.98 294.40 343.70 326.93 316.14 354.03 409.25 393.20
Fallow 216.67 225.19 245.05 204.22 257.67 272.86 299.54 260.81
Conv. 204.07 224.62 224.62 177.96 250.61 255.25 276.91 228.08
LSD0.05 for comparing variable cost means among planting dates = 2.47
LSD0.05 for comparing variable cost means among tillage systems = 2.47
LSD0.05 for comparing total cost means among planting dates = 2.47
LSD0.05 for comparing total cost means among tillage systems = 2.47
Table 6. Influence of planting date and tillage system on
average net returns.
Tillage Planting date
system April May June July
No-till 412.27 358.67 164.55 -25.34
Fallow 464.28 490.91 218.37 56.71
Conv. 453.93 449.52 240.31 13.81
LSD0.05 for comparing among planting dates 11.65
LSD0.05 for comparing among tillage systems 12.71
Soil Water, cm
April May June July Rainfall
April May June July August
Fig 1. Influence of planting date and rainfall on soil water storage to a depth of 60 cm
at Keiser in 1996. LSD (a) for comparing between planting dates. LSD (b) for
comparing between sample dates.
Rainfall, cm
E.C. Gordon, T.C. Keisling, L.R. Oliver and Carl Harris 1
Anecessary situation that occurs in the cotton gin-
ning process is the accumulation of about 200 lb
of waste per ginned bale. This waste, called gin
trash, has to be disposed of at some point in time. Much
of the gin trash was incinerated for many years, but cer-
tain regulations, such as the Clean Air Act of 1970, have
removed burning as an option. Using gin trash as a live-
stock feed is done to an extent, but there is some concern
regarding chemical residues.
Another option in the disposal of gin trash is to spread
it directly on the fields. Returning the organic material
and nutrients can be beneficial, but certain problems might
occur when spreading raw gin trash onto fields. Weed
seed and disease, particularly Verticillium wilt, may be
introduced to or increased in fields when spreading raw
gin trash. The removal of these two potential problems
makes the spreading of gin trash much more attractive.
An effective method of handling gin trash and reducing
the problems associated with weed seed and disease or-
ganisms is to compost the material. With adequate mois-
ture, approximately 70%, the heat generated in the
composting process can be sufficient to kill weed seed
(140 F for 10 days) and disease organisms (145 F for two
days) (Alberson and Hurst, 1964; Griffis and Mote, 1978b;
Parnell et al., 1980). Commercial contained-composting-
systems have demonstrated this. However, the high cost
of commercial contained-composting-systems tends to be
prohibitive, so alternative composting methods have been
Windrow-composting-systems can generate the neces-
sary heat if there is adequate volume, moisture and aera-
tion. The aeration is usually provided by turning the
composting material with some type of implement. In the
humid Southern region, rainfall could conceivably supply
sufficient water for initial wetting of the gin trash as well
as keeping it moist for the duration of the composting
process. This would eliminate a wetting step and make the
overall process cheaper.
Recently, new gin trash handling methods have been
developed. The Lipsey®-gin-trash-composting-system re-
1Research Associate, NEREC, Keiser, AR; Professor , NEREC, Keiser, AR;
University Professor, Dept. Of Agron, Univ. Of Ark., Fayetteville, AR; and
County Ext. Agt. Deceased.
quires the compost to stay in place. The compost pile is
formed in a circular pattern by rotating back and forth
around a pivot point (Fig 1. top view). The rotation motion
is at a constant speed so the thickness of gin trash depos-
ited on top of the compost pile is a function of 1) amount
of trash in un-ginned cotton, 2) rate of ginning and 3)
current depth of compost pile (as the sides are slanted as
shown in Fig. 1 side view). Uniform wetting throughout
the pile is facilitated by wetting the gin trash as it is deliv-
ered to the top of the compost pile. The resulting com-
post pile has layers of various thicknesses that are applied
at varying rates. Thus, the zone of aeration is controlled
by the depth from an outside surface and the duration of
the compost at this depth.
Experiments were conducted to evaluate certain aspects
of windrow-composting-systems and the Lipsey® system.
Experiment 1
In March 1977 gin trash from Mann’s Gin in Lee
County, Arkansas, was placed in windrows for composting.
A typical windrow is approximately 40 ft long, 4.5 ft at
the base, 2 ft across the top and 1.33 ft tall. The experi-
mental design was a randomized complete block with five
replications. The treatment design was a split-split plot.
The main plots were timing of turning of the windrow with
a root rake. Main plot treatments included 1) turned weekly
or 2) turned when the temperature 6 in. below the surface
reached 80 F. Main plots were split with half receiving
4.2 lb of nitrogen (N) per plot as a commercial fertilizer
and the other receiving no N. The N-treated plots were
then split and one-half of each plot inoculated with
RoebicTM aerobic inoculum. Temperatures at 6 in. from
the windrow top surface were taken daily until mid-April
when composting was complete and were used to evaluate
the benefit of additives in the composting process. Rain-
fall was the only water received by the compost piles.
Composite samples were collected before and after
composting and analyzed for nutrients and selected chemi-
Experiment 2
Composting plots were established in Lee County, Ar-
kansas, during November 1978 to evaluate aeration meth-
ods. Two implements were compared for effectiveness of
turning a windrow–a root rake and a modified combine
(Lalor et al., 1978). The experimental design was a ran-
domized complete block with four replications. Treat-
ments consisted of turning the compost weekly and every
two weeks by each machine. Moisture was monitored.
Those plots turned with the combine had water added to
the compost pile to adjust moisture to circa 70%. The
plots turned with the root rake received only rainfall for
wetting the compost pile. Effectiveness was determined
by measuring internal temperatures as in Experiment 1.
Experiment 3
In February 1995 three gin trash compost piles that
were formed during the fall of 1994, using the Lipsey®
gin trash composting system, were selected for sampling
and evaluating weed seed germination. Two compost piles
were located in Phillips County, Arkansas, and one in
Crittenden County, Arkansas. In Phillips County samples
were taken from both piles to a depth of 30 in. in 6-in.
increments from the surface using a bucket. Approximately
2.5 gallons of compost was removed from each depth
increment in each pile for subsequent analysis for chemi-
cals and organisms.
The compost pile in Crittenden County was sampled
using a front-end bucket loader to cut into the pile 10 to
15 ft. Again, approximately 2.5 gallons of compost was
collected at 5-in. increments from the compost surface to
a depth of 48 in. for subsequent analysis by grabbing ma-
terial from an 8-ft-long vertical face.
All samples from each location were stored in plastic
bags and kept at room temperature until they were taken
to the University of Arkansas at Fayetteville within one to
two weeks after collection. The samples were divided into
two sub-samples of approximately 1 gallon each. The sub-
samples were placed in containers measuring 16 in. long
by 12 in. wide by 2 in. tall. The containers were placed in
a greenhouse for 10 weeks. The compost in each con-
tainer was kept moist and stirred every two weeks. Obser-
vations were made two to three times weekly on the num-
ber and weed species that germinated. Chemical composi-
tion was determined for N, C, P, K, Ca and Mg. The pH
was also measured.
Experiment 1
Neither use of a starter aerobic inoculum (Fig. 2) nor
addition of N (Fig. 3) was needed to initiate the composting
process. Regardless of treatment, temperatures were simi-
lar over the composting period. This indicated that no ad-
dition of inoculum or N was needed for proper composting
to occur. These findings agree with those of Griffis and
Mote (1978a). Heating criteria for turning the pile gave
slightly higher internal temperature than just turning weekly
(Fig. 4). Neither method resulted in temperatures high
enough or long enough to kill Verticillium wilt fungi or
weed seeds. Weeds observed growing on top of the wind-
rows after composting was complete were annual blue-
grass (Poa annua), large crabgrass (Digitaria
sanguinalis), purple nutsedge (Cyperus rotundas), yel-
low nutsedge (Cyperus esculentus), pigweed (Amaranthus
spp.), morningglory (Ipomoea spp.), horsenettle (Solanum
carolinense) and prickly sida (Sida spinosa). Reproduc-
tive characteristics of certain weeds listed above make it
obvious that the seed were mixed with the compost during
the turning process rather than being delivered in the gin
The nutrient analysis of gin trash samples is shown in
Table 1 as total analysis and soil test analysis. The pH
levels remained below 7, indicating aerobic composting
conditions. Higher pH levels would indicate anaerobic
composting that favors the conversion of N to ammonia.
High temperatures enhance the volatility of ammonia
(Golueke, 1972).
We observed that using natural rainfall for wetting re-
sulted in channeling of water through selective pathways
in the compost pile. As a result, some of the material was
very slow in wetting and did not necessarily go through a
heat. These pockets of dry material were mechanically
incorporated with wetter compost during the turning pro-
Experiment 2
Due to the non-uniform wetting, a modified combine
(Lalor et al., 1978) that would mix and wet a windrow
uniformly was built. The modified combine accelerated
the composting process, as evidenced by increased early
composting temperatures (Fig. 5). The temperatures were
still not high enough or long enough to kill weed seeds
and wilt organisms 6 in. below the compost pile surface.
Weekly mixing moves materials from the outside of
the compost pile to the inside where heat can be accumu-
lated. This should result in temperatures high enough and
long enough in duration (140 F for 48 hr) to kill Verticil-
lium wilt organisms and weed seeds between weekly turn-
ings. Assuming that 50% of the pile is wet enough to
generate sufficient heat, complete weekly mixing provides
sufficient aeration and carbon supply for the composting
organisms to function. After each mixing, the reduction in
the compost volume containing viable diseases and weed
seeds should be halved. Therefore, 15 mixes or weeks
would be required to produce a 99.99% compost with
essentially no weeds or diseases, which is about twice as
long as it took our composting operation to be completed.
Hence, a different method other than windrow-composting
with mechanical mixing will be necessary.
Experiment 3
No viable weed seeds were detected from the compost
samples obtained from the compost piles made by the
Lipsey® composting system. Two months after the gin-
ning season, temperature within the pile was too hot for
more than 10 to 15 seconds contact with the bare hands.
Since no weed seeds germinated in our greenhouse test, it
appears that the weed seed viability was destroyed from
the heat of composting. The outside of the pile, which had
not gone through a heating process, had several weeds
growing on it. This might be easily sterilized using a solar
technique, such as covering the entire pile with a sheet of
black plastic for a few days.
The carbon/nitrogen ratio (C/N) tends to increase at
the deeper sampling depths, indicating an anaerobic
composting process with a possible loss of N as ammonia
(Table 2). The pH levels being greater than 7 at depths
greater than 6 in. confirm anaerobic conditions. The
anaerobic composting process appears to generate suffi-
cient heat for sterilization but will result in a compost of
a fibrous consistency with a bad odor.
The results presented here indicate that the windrow
composting system does not solve the two problems of
Verticillium wilt or weeds associated with gin trash. Oth-
erwise, the compost obtained is quite satisfactory as a
product. The Lipsey® composting system produced a com-
post pile whose outer layer had problems with weed seed
and Verticillium wilt survival. These problems could be
easily eliminated by using a solar sterilization process
consisting of covering the pile with a continuous sheet of
plastic. Otherwise, the composting process turns anaero-
bic within a couple of feet of the surface, resulting in
incomplete composting and in offensive odors.
Alberson, D.M., and W.M. Hurst. 1964. Composting cotton gin waste.
USDA-ARS, Publication ARS 42-102.
Golueke, C.G. 1972. Composting: A study of the process and its
principles. Rodale Press, Inc.
Griffis, C.L., and C.R. Mote. 1978a. Cotton gin trash composting
studies. Arkansas Farm Research 27(4):3.
Griffis, C.L., and C.R. Mote. 1978b. Weed seed viability as affected
by the composting of cotton gin trash. Arkansas Farm Research
Lalor, W.F., D. Berry, C. Harris and J.K. Jones. 1978. Compost-
making equipment for cotton gins. ASAE paper no. 78-3545.
Parnell, P.E., E.R. Emino and E.K. Grubaugh. 1980. Cotton gin trash:
can it be safely utilized? Agricultural Engineering 61:21-22.
Table 1. Analysis of gin trash used in 1977 experiments.
Total Analysis N P K Ca Mg Na Zn Fe Mn As pH
-----------------------------------------%--------------------------------------- ----------------------ppm---------------------
Before Composting 1.66 0.29 0.78 1.90 0.34 0.05 41.0 2218 343 2.0 6.9
After Composting 1.04 0.14 0.52 0.51 0.21 0.02 -1280 313 - 6.2
Soil Test Analysis Nitrate-N P K Ca Mg Na EC pH
------------------------------------lb/acre---------------------------------- µmhosx103
1620 160+ 2740 1699 710 193 1.4 6.2
Table 2. Chemical analysis of compost from a Lipsey® system for handling gin trash
in Phillips County (PC) and Crittenden County (CC), Arkansas.
Depth from N C C/N P K Ca Mg pH
in. -------------------------------------------------------------------------%----------------------------------------------------------------------------
0-6 4.0 4.0 26.70 30.2 6.8 7.5 0.6 0.4 0.6 0.5 2.6 2.2 0.6 0.4 6.2 5.6
6-12 4.5 4.0 29.20 13.7 6.4 6.2 0.7 0.6 2.1 2.4 2.5 2.2 0.8 0.6 7.0 7.8
12-18 3.9 3.9 29.70 31.1 7.7 8.0 0.8 0.6 1.6 1.9 2.8 2.2 0.7 0.6 7.4 7.5
18-24 4.0 3.9 32.40 27.1 8.0 6.9 0.6 0.6 1.6 2.3 2.9 2.3 0.6 0.6 8.0 7.7
24-30 4.0 4.1 31.50 29.7 8.0 7.2 0.6 0.5 2.3 1.7 2.7 1.7 0.7 0.5 7.3 6.9
30-36 3.9 3.5 36.50 28.7 9.4 8.3 0.6 0.6 2.8 1.9 2.5 2.9 0.7 0.6 7.0 7.7
36-42 - 3.3 - 34.1 -10.5 -0.6 -1.8 -2.9 -0.6 -7.3
42-48 - 2.7 - 36.1 -13.1 -0.5 -1.9 -2.7 -0.5 -7.6
E.M. Holman, A.B. Coco and R.L. Hutchinson1
Advances in equipment and herbicide technology
have contributed greatly to the increase in pro-
ducer acceptance of reduced tillage practices in
northeastern Louisiana. Reduced soil erosion (Hutchinson
et al., 1991), increased soil organic matter (Boquet and
Coco, 1993) and reduced soil moisture evaporation
(Wilhelm et al., 1986) are just some of the documented
benefits from no-tillage. Reduced tillage, in many instances,
has also led to lower equipment and fuel costs and savings
in time and labor (Laws, 1993). In addition, cover crops
have been found to be an important component of conser-
vation tillage systems (Hutchinson et al., 1991; Ebelhar et
al., 1984).
Although erosion is not a serious problem on many of
the clay soils in the Mississippi River Delta, cotton
(Gossypium hirsutum L.) production has still benefitted
from reduced tillage practices primarily by allowing pro-
ducers to plant in a more timely fashion (Boquet and Coco,
1993). Spring tillage on clay soils often results in a cloddy,
dry seedbed in which it is difficult to obtain a uniform
plant stand.
On clay soils, deep tillage to relieve compaction has
traditionally been considered unnecessary due to the natu-
ral shrinking and swelling that these soils undergo as the
moisture content cycles from wet to dry. It has been specu-
lated (Smith and Whitten, 1992) that while clays do not
develop compaction pans typical of lighter-textured soils,
they may develop compacted blocks of soil beneath the
plow layer. The effect of this soil condition is to confine
plant roots to the soil volume near the block surfaces. The
density of the blocks prevents or severely restricts root
growth into the clay block, and roots that do grow from
one block surface to another are often broken when the
blocks dry and shrink. Results from previous tillage stud-
ies failed to demonstrate crop response to deep tillage on
clay soil (Raney et al., 1954; Saveson et al., 1958; Tupper,
1978; Heatherly, 1981). However in these studies, the
tillage operations were performed in the spring when the
subsoil was most likely wet from winter rainfall. Recently,
Smith (1995) indicated that deep tillage in the fall, when
the soil profile was dry, was beneficial for cotton growth
and yield on a Tunica clay. There is a lack of information
1Assistant Prof., Louisiana State University Agricultural Center, Northeast
on the interaction between deep tillage in the fall and
various conservation tillage practices on clay soils in Loui-
On the medium- and coarse-textured alluvial soils in
northeastern Louisiana, compaction is a yield-limiting fac-
tor unless some form of deep tillage is performed
(Crawford, 1979; Saveson et al., 1958). In northeastern
Louisiana, these soils have typically been under a mono-
crop production system that utilizes extensive surface till-
age to control weeds, prepare seedbeds and incorporate
herbicides. Although these soils are highly productive, the
combination of extensive tillage and mono-crop culture
have contributed to low organic matter levels (< 1.0%) in
many fields. As the use of conservation tillage practices
and winter cover crops has been shown to result in in-
creases in soil organic matter levels (Hutchinson et al.,
1991; Millhollon and Melville, 1991), some combination
of these practices might lead to improved growth and yield
of cotton on these soil types. Therefore, the objectives of
this study were to: 1) determine the optimum combina-
tion of cover crop and tillage necessary to maximize cot-
ton production while maintaining or increasing soil pro-
ductivity and 2) examine the effect of deep tillage in con-
junction with cover crops and reduced tillage practices on
cotton production.
A field study was initiated in the fall of 1996 on a
Commerce silt loam (fine-silty, mixed, nonacid, thermic
Aeric Fluvaquent) and on a Sharkey clay (very-fine, mont-
morillonitic, nonacid, thermic, Vertic Haplaquepts) at the
Northeast Research Station near St. Joseph, Louisiana. A
total of 16 treatments were established with combinations
of tillage systems {no-till (NT), conventional till (CT),
reduced-till (RT)), winter cover crops [winter wheat (Triti-
cum aestivum L.), hairy vetch (Vicia villosa L.), and na-
tive vegetation], in-season cultivation, and fall sub-soil-
ing} summarized in Table 1. Treatments were only slightly
different on the two soil types with CT on the silt loam
including disking in the fall and spring prior to seedbed
preparation, while on the clay CT involved only hipping in
the fall and spring. The RT treatments on the silt loam
were hipped in the fall and spring, while on the clay the
RT treatment involved hipping and rolling in the fall and
no additional tillage in the spring. Experiment design for
both tests was a randomized complete block with four
Research Station, St. Joseph, LA.
replications. Plot size was four rows (40-in. row spacing)
by 65 ft.
Deep tillage operations on the appropriate plots were
conducted with a Paratill following cover crop planting in
October 1996. Cotton cultivar ‘Suregrow 501’ was planted
5 May 1997 using ripple coulters mounted on the planter.
Management of the cover crops prior to planting (4 weeks
before planting) in the no-till plots consisted of 1) an
application of glyphosate (1.0 lb ai/acre) followed by
paraquat (0.75 lb ai/acre) on the wheat plots; 2) an appli-
cation of paraquat (1.0 lb ai/acre) and cyanazine (0.75 lb
ai/acre) followed by paraquat (0.75 lb ai/acre) on the vetch
plots; 3) an application of paraquat (0.75 lb ai/acre) and
cyanazine (0.75 lb ai/acre) on the native plots. Preemer-
gence weed control in all plots consisted of a broadcast
application of pendimethalin (1.0 lb ai/acre) plus
fluometuron (1.2 lb ai/acre). All appropriate NT, CT and
RT treatments were cultivated twice. Additional herbicide
applications included broadcast application of pyrithiobac
(0.079 lb ai/acre), post-directed application (banded) of
prometryn plus MSMA (0.31 and 1.0 lb ai/acre) and a
layby application (broadcast) of cyanazine and MSMA (1.1
and 1.65 lb ai/acre).
Based on past work with these cover crops, nitrogen
fertilization of the cotton was adjusted to 60 lb N/acre
following vetch, 120 lb N/acre following wheat with the
other plots receiving 90 lb N/acre. The middle two rows
were harvested from each plot 17 October with a spindle
picker. On 22 October 1997, following cotton stalk de-
struction, the wheat and vetch cover crops were planted in
the respective plots. The next day, treatments were split
for deep tillage using a Paratill, and the appropriate treat-
ments were disked or hipped.
All data were analyzed using the ANOVA or GLM pro-
cedures of SAS (SAS Institute, 1989). In order to assess
individual treatment factor effects, contrast statements
were used following the GLM procedure.
Two days after planting, soil temperature was lower (3
to 5 F) in the furrow (2-in. depth) in the vetch plots com-
pared to the conventional or reduced tillage treatments on
both soil types (data not shown). Although the wheat plots
were numerically lower than the conventional plots, the
differences were not significant. The differences in soil
temperature could help to explain some of the observed
differences in early growth.
Commerce Silt Loam
Nodes above white flower (NAWF) was affected by
some of the treatment factors. With regard to NAWF val-
ues recorded on 30 July, the no-till plots had a higher
value than conventional or reduced tillage treatments (5.2
vs 4.6 or 4.7), indicating a slight delay in maturity. On the
same date there were no differences in NAWF between
no-till plots with respect to cultivation (5.2 vs 5.2). Within
the no-till plots at this date, cotton in vetch treatments
was later maturing than cotton in the plots with a wheat
cover crop or native cover (5.4 vs 5.1 or 5.1). This could
be partially explained by the lower early soil tempera-
tures, which could have reduced early plant vigor. There
was also a difference in NAWF at this date between the
plots that were sub-soiled and the plots that were not (5.2
vs 4.9).
Cotton yield was also affected by some treatment fac-
tors; contrast statements were again used in order to ex-
amine the influence of individual treatment variables. There
was no difference in yield between the no-till treatments
and the conventional or the reduced till treatments. There
was also no difference in yield between the no-till plots
that were cultivated and those that were not. With respect
to the cover crops, there was no difference between the
wheat and the native treatments. However, both the wheat
and the native were higher than the vetch treatments (2641
and 2651 vs 2470 lb seedcotton/acre). This could be re-