Bioresource Technology 96 (2005) 47–53

Processing cotton gin trash to enhance in vitro dry matterdigestibility in reduced time

Lester O. Pordesimo a,*, Samuel J. Ray a, Michael J. Buschermohle a,John C. Waller b, John B. Wilkerson a

a Department of Biosystems Engineering and Environmental Science, The University of Tennessee, 2506 E.J. Chapman Drive Knoxville,

TN 37996-4531 USAb Department of Animal Science, The University of Tennessee, 2505 River Drive, Knoxville, TN, 37996-4500 USA

Received 30 September 2002; received in revised form 23 February 2004; accepted 23 February 2004

Available online 13 April 2004

Abstract

Cotton gin trash (CGT) in the raw form is poorly digested by ruminants due to lignocellulosic complexes. These structures must be

broken down before adequate digestion can occur. This may be performed by physical and/or chemical means. Two methods for

improving digestibility are particle size reduction and/or treatment with sodium hydroxide (NaOH). To evaluate the effectiveness of

each method, three experiments were performed in which different CGT types were tested. Each type represented trash from a

particular cleaning stage in the cotton ginning process. First, each type was ground with a knife-type grinding mill using screen sizes

0.5, 1.0, and 2.0 mm. For the second experiment, particle size was held constant at 2 mm, and all CGT types were treated with 4% and

6% NaOH (w/w) at room temperature. An agitation cycle of 5 min on and 10 min off was used, with the total mixing time being 4 h.

Lastly, particle size and NaOH concentration were held constant, and treatments were performed at room temperature, 40 �C, and50 �C. The total mixing times were 2 and 3 h for 50 and 40 �C, respectively. For all experiments two subsamples of each treatment were

tested for in vitro dry matter digestibility (IVDMD). From grinding alone, digestibility increased as particle size decreased. Grinding

to 0.5 mm resulted in an average IVDMD of 47.8% while grinding to 2.0 mm resulted in an average IVDMD of only 33.8%.

Digestibility also improved with a greater NaOH concentration. An average in vitro digestibility of 70.5% was achieved with 6%

NaOH (w/w) treatment, essentially doubling that of the raw CGT. Increasing the reaction temperature did not result in increased

digestibility because the mixture dried out, with a consequent reduction in chemical distribution and uniformity in heat transfer. There

are still chemical residues in the CGT, and elimination/reduction of these is an issue that needs to be addressed in further research.

� 2004 Elsevier Ltd. All rights reserved.

1. Introduction

Finding an economical and practical method of dis-

posing of the large quantities of cotton gin trash (CGT)

that accumulate during the ginning process remains aproblem for cotton ginning facilities. In the past,

incineration has been a widely used disposal method,

but environmental regulations have practically elimi-

nated this method. Using CGT as a ruminant feed

ingredient is one potential high-value option for dis-

posing of the material in an environmentally friendly

manner. Comparatively, CGT has lower feed value than

alfalfa, but higher than prairie hay, bermuda hay, rice

*Corresponding author. Tel.: +1-865-974-7266; fax: +1-865-974-

4514.

E-mail address: pordesim@utk.edu (L.O. Pordesimo).

0960-8524/$ - see front matter � 2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.biortech.2004.02.031

hulls, and sorghum stover, according to Lalor et al.

(1975) (Table 1).

Problems related to utilization of CGT as a livestock

feed/livestock feed ingredient include chemical residues,

digestibility, and nutritive value (Buser, 1999). Whilestill a problem, the issue of chemical residues is not as

severe as in the past. Levels have decreased since the

1970s, and current residues are relatively short-lived

(personal communication from W.D. Mayfield cited by

Thomasson et al. (1998)). All these issues may poten-

tially be addressed and ameliorated with a reduced

processing time by engineering physical, chemical, and

enzymatic pretreatment and nutrient blending processesfor CGT. The strategy is to quickly pre-digest a fraction

of the lignocellulosic material in cotton gin trash to

facilitate a more complete digestion and utilization by

ruminants, and to increase protein content through

Table 1

Nutritive value of cotton gin trash

Nutritional component Analysis

Estimated net energy maintenance (MCa/Cwt) 46

Estimated net energy production (MC/Cwt) 15

TDN ruminant 45

Crude protein (%) 7.0

Crude fat (%) 1.5

Crude fiber (%) 35.0

Ash (%) 10.0

Calcium (%) 0.15

Phosphorous––Total (%) 0.09

Potassium (%) 0.87

Source: Lalor et al. (1975).aMC¼million calories.

48 L.O. Pordesimo et al. / Bioresource Technology 96 (2005) 47–53

blending with other ingredients. Such a strategy requires

characterizing the physico-chemical properties of the ginwaste (broadly classified as picker or stripper trash), and

then investigating the main effects and interactions of

such pretreatment process variables as particle size,

chemical concentrations, cycled agitation, and reaction

time and temperature. While there have been many

studies involving the chemical treatment of crop residues

for enhanced livestock feeding value, there is little

information on the effects of treatment process variablesother than chemical concentration.

Han et al. (1983) stated that in cellulose hydrolysis,

direct physical contact between the hydrolytic agents and

the cellulose molecules is required and that the rate of

hydrolysis is a function of the cellulose surface area

accessible to the hydrolytic agent. This translates to im-

proved digestion of lignocellulosic crop residues with a

decrease in the particle size of the material. Working withrice straw and bagasse, Agrawal et al. (1989) studied the

effect of particle size on the in vitro digestibility. How-

ever, no published research was found that directly

investigated the effects of size reduction of the various

CGTmaterial (and crop residues in general) on ruminant

digestibility. It has also been proven that treating CGT

with sodium hydroxide (NaOH) increases rumen turn-

over (Conner and Richardson, 1987), but there havebeen no reported studies on the chemical pretreatment of

segregated CGT types (materials separated by particular

cleaning stages in the cotton ginning process), much less

the use of agitation and elevated process temperatures to

achieve effective pretreatment over a limited time. Dry

matter digestibilities of crop residues range from 35% to

60%, with corn and sorghum plant material generally

higher than small grain straws (Klopfenstein, 1980).Thomasson (1990) indicated that raw cotton gin trash

had an in vitro digestibility of 34%, which falls in the

lower end of crop residue digestibility range.

The objective of this preliminary study was to deter-

mine the impact of particle size reduction and agitated

treatment with aqueous NaOH solution on the in vitro

dry matter digestibility (IVDMD) of cotton gin trash,

with the goal of reducing treatment time. Both of these

methods should aid in hastening the breakdown of the

lignocellulosic complexes, thereby enhancing ruminant

digestibility. Factors such as convenience, time econ-

omy, effectiveness, and cost are critical for a pretreat-

ment process to be commercially viable for ginners.

Recognizing this, an essentially dry process was studied:

liquid delivery into dry cotton gin trash. More recentstudies concerning the degradation of biomass have

been liquid systems. It is hypothesized that an appro-

priate combination of process variables can result in

acceptable improvement in digestibility over a 4-h

treatment period. The effects of CGT type, particle size,

chemical concentration, and processing temperature

were also investigated.

2. Methods

Various types of CGT from the ginning of stripper-

harvested and spindle-picked cotton were obtained from agin in West Tennessee. Each type represented trash from

a particular cleaning stage in the cotton ginning process. A

consolidated sample of CGT cleaned from stripper-har-

vested cotton was also obtained from Texas. Listed in

Table 2 are different CGT types tested and their moisture

contents, determined following ASTM Standard D 2495

(ASTM, 1987). Types A, B and F, G were mostly non-lint

material whereas types D, E and I, J were mostly lint. Thetest materials were kept in large plastic bags and stored in

an air-conditioned laboratory (�25 �C) before treatment

three months after they were obtained.

Three two-factor experiments were performed in

which all CGT types were tested. The effect of particle

size and CGT type on IVDMD was evaluated in the first

experiment. In the second experiment, the effect of

NaOH concentration and CGT type on IVDMD wasevaluated while maintaining a constant particle size. In

the third experiment, NaOH concentration and particle

size were held constant while testing the effect of dif-

ferent process temperatures and CGT type on IVDMD.

Material, time, and cost constraints limited the studies

to two replicates. IVDMD was evaluated for all samples

using the Moore modification (Moore et al., 1972) of the

Tilley and Terry (1963) two-stage technique. Statisticalanalysis for the effect of the experimental treatment

factors was accomplished through analysis of variance

using the Mixed procedure of the Statistical Analysis

System (SAS Institute, Inc., 1999). Comparison between

treatment means and trend comparisons were investi-

gated using orthogonal contrasts. Specific contrasts

accomplished compared the IVDMD between stripper-

harvested and spindle-picked cotton and high- and low-lint CGT types. The effective trend in IVDMD caused

by a progression in particle size, NaOH concentration,

or process temperature was also evaluated.

Table 2

Cotton gin trash types tested

Code Description MC (% w.b.)

Stripper-harvested cotton––West Tennessee

A First stage cleaning––inclined cleaner (essentially burr and stick separator) 9.60

B Second stage cleaning––inclined cleaner 9.66

C Material underneath gin stand 8.21

D First stage lint cleaner 7.17

E Second stage lint cleaner 5.65

Spindle-picked harvested cotton––West Tennessee

F First stage cleaning––inclined cleaner (essentially burr and stick separator) 8.99

G Second stage cleaning––inclined cleaner 8.75

H Material underneath gin stand 7.26

I First stage lint cleaner 6.64

J Second stage lint cleaner 6.01

Stripper-harvested cotton––Texas

M Mixture of all cleanings 8.76

L.O. Pordesimo et al. / Bioresource Technology 96 (2005) 47–53 49

2.1. Experiment 1

A completely randomized design with split-plot

treatment arrangement was used to study in vitro dry

matter digestibility: CGT type was in the main plot

while particle size was in the sub-plot. The bulk

material of each CGT type procured was split and

randomly designated as replication 1 or 2 [model:y ¼ meanþ typeþ repðtypeÞ þ sizeþ size � typeþsize � repðtypeÞ�. The duplicate samples of each type

were then ground in a knife mill (Model P15, Glen

Mills, Inc., Clifton, N.J., USA) using screen sizes 0.5,

1.0, and 2.0 mm in a randomized manner. Samples

comprising the various type-particle size treatment

combinations were then evaluated for in vitro dry

matter digestibility.

2.2. Experiment 2

CGT types ground through a 2.0 mm screen wereused as the test material for this experiment. Since

greater energy is required to grind material to smaller

particle sizes, it was more practical to undertake chem-

ical treatment testing with the largest of the test material

sizes. Each type was treated with 4% and 6% NaOH

(w/w) on a dry matter basis. These concentrations used

were based on recommendations from Han et al. (1983)

using ryegrass as a test material. Additional support forthese choices came from Klopfenstein (1978), who re-

ported optimum animal response to NaOH at levels

from 3% to 5% of residue dry matter.

Chemical treatment was achieved using a Kitchen

Aid Classic stand mixer (KitchenAid, Greenville, Ohio,

USA) to provide agitation and greater interaction be-

tween the NaOH solution and the CGT. Four mixers

were used simultaneously to reduce the time to accom-plish the chemical treatment experiments. For optimum

mixing effectiveness, a minimum level of fill of material

was required in the mixer bowl. Filling the bowl to one

third of its total depth, approximately 8 cm from the

bowl bottom, provided for effective mixing. Each test

was thereby started by pouring ground CGT up to the

designated fill level. Since the CGT had different densi-

ties, the total mass of material treated was different for

each test. Dry matter of each mixer load was calculated,and the amount of NaOH to produce a 4% or 6%

treatment was then calculated and weighed. The NaOH

was added to the CGT as an aqueous solution. The

amount of water used in the solution was determined by

observing the effectiveness of mixing in several test

mixes. Tests showed that a raw material:water ratio of

1:2 was adequate to homogeneously transport the

NaOH, but not too much as to make the mixture aslurry rather than a moist solid. Thus, a solid to liquid

ratio of 1:2 was set for the experiments to minimize the

amount of water needed for the treatment process and

minimize the amount of wastewater generated in a

possible commercial process. The required amount of

pure NaOH was mixed in a beaker with an amount of

distilled water equal to twice the weight of the CGT

already in the mixer bowl. Since the weight of the solidNaOH was not considered in determining the weight of

water needed, the solid:liquid ratio was actually just

fractionally short of being 1:2. The mixer was set on low,

operated for 4 h using a cycle of 5 min on and 10 min off.

The pretreated samples were transferred to airtight

sample bags and kept in frozen storage ()17 �C) until allexperimental treatments were completed. Samples were

then evaluated for in vitro dry matter digestibility. Theexperiment was conducted as a completely randomized

design with factorial treatments ½model: y ¼ meanþtypeþNaOHþ type � NaOHþ repðtype � NaOHÞ�.In undertaking statistical analysis, the results for CGT

ground to 2 mm in Experiment 1 were used as the no-

chemical control.

50 L.O. Pordesimo et al. / Bioresource Technology 96 (2005) 47–53

2.3. Experiment 3

The 11 CGT types ground through a 2.0 mm screen

were also used as the test material in this experiment.

Each type was treated with 6% NaOH following

Experiment 2 procedures with the use of temperature

during mixing as a treatment variable [ambient

(�25 �C), 40 �C, 50 �C]. Electric rope heaters (OmegaluxFR100, Omega Engineering, Inc., Stamford, Conn.,

USA) were coiled around the base of the mixing bowls

to provide heat. The power for the rope heater was

regulated in order to maintain the targeted elevated

temperatures. To prevent the material from charring on

the bowl surface during processing, heat was applied

only when agitation occurred. Temperature measure-

ments were taken at the end of each mixing/heatingstage with an infrared thermometer (Model no. 39650-

20, Cole-Parmer Instruments, Co., Chicago, Ill., USA)

to verify that the targeted temperatures were achieved.

Since the water evaporation rate increases with tem-

perature, the total mixing time had to be reduced to 3

and 2 h for 40 and 50 �C, respectively. Mixing time was

retained at 4 h for pretreatment at room temperature.

By adjusting the process times at each of the processtemperatures tested, these two factors became con-

founded with each other. The experiment was conducted

following a completely randomized design with factorial

treatments ½model: y ¼ meanþ typeþ tempþ type �tempþ repðtype � tempÞ�. Pretreated samples were

handled as in Experiment 2. Samples comprising the

various type-process temperature treatment combina-

tions were thereafter evaluated for in vitro dry matterdigestibility.

2.4. In vitro dry matter digestibility

Dry matter digestibility was determined using theMoore modification of the Tilley and Terry (1963) two-

stage in vitro technique (Moore et al., 1972). The donor

animal was maintained on a medium quality alfalfa

(Medicago sativa) diet with supplementation of 0.454

kg/d of 48% of soybean meal. The supplementation was

given at 0700 with collection being 2-h post-feeding.

Whole ingesta from the top half of the rumen were re-

moved and transferred to an insulated container thenstrained through four layers of cheesecloth. Fluid was

transported from The University of Tennessee Dairy

Farm to the nutrition laboratory (about 15 min) where it

was flushed with CO2 and then placed into a large glass

bottle in the water bath containing the McDougall’s

saliva (McDougall, 1948). An appropriate quantity of

McDougall’s saliva was prepared prior (40 ml per tube

plus 200 ml extra) and placed into a 13.25 l bottle. Oneml of 4% CaCl2 per liter of McDougall’s saliva was

added to the bottle and then it was placed into a 39 �Cwater bath and CO2 was bubbled gently through the

bottle. The pH of the buffer was adjusted to between 6.9

and 7.0 in order to mimic the natural buffer in the ani-

mal more closely. One part rumen fluid was added to 4

parts buffer and was allowed to mix for 10 min by action

of the bubbling CO2.

Five grams of each sample was weighed and placed

into a numbered centrifuge tube. Two ml of distilled

water were added to each tube to moisten the sample.Each tube was agitated with a test tube mixer. Fifty ml

of media were gently added to each centrifuge tube with

using an automatic pump. The tubes were placed into a

water bath immediately after adding the media and

flushed with CO2 for 15 s. The tubes were sealed quickly

with a rubber stopper fitted with a Bunsen valve. The

tubes were placed in a 39 �C water bath and swirled 1 h

later to ensure that all of the sample particles weresoaked with media. This swirling was repeated twice the

first day and thrice the second day.

After 48 h of incubation, the rubber stoppers were

removed and the sample particles adhering to the stop-

pers were washed into the tubes with distilled water. One

ml of 20% HCl was added to each tube and swirled and

then repeated (total 2 ml per tube). An additional 4 ml

of HCl was added to each tube and then swirled. Two mlof 5% pepsin were then added to each tube followed by

swirling to mix the pepsin with the acid and sample. The

rubber stoppers were replaced and the tubes returned to

the 39 �C incubator. The swirling was repeated twice the

first day and thrice the second day.

Gooch crucibles were then prepared by forming a

glass wool mat in the bottom to serve as a filter. After

46 h of pepsin digestion, the contents of the centrifugetubes were transferred to the Gooch crucibles. All

residual material from the tubes was rinsed out with hot

distilled water and the Gooch crucibles were rinsed with

hot distilled water. The Gooch crucibles were placed

into a drying oven overnight at 105 �C. The following

day they were removed from the drying oven, cooled in

a desiccator, and weighed.

3. Results and discussion

3.1. Particle size

As illustrated in Fig. 1, IVDMD increased as particle

size decreased ðP < 0:01Þ, which was expected. Smaller

particle sizes have a larger total surface area for the

action of the digestive fluids. Grinding to 0.5 mm re-

sulted in the highest IVDMD of the three particle sizes

evaluated, with an average digestibility of 47.8%.

Grinding to 2.0 mm resulted in the lowest digestibility of

33.8%. It should also be noted that there were consid-erable differences in digestibility between the same sizes

for the different CGT types. Generally, regardless of

how the cotton was harvested, the types with higher fi-

0102030405060708090

100

A B C D E F G H I J MCGT Type

IVD

MD

, %

0.5 mm1.0 mm2.0 mm

spindlestripper

Fig. 1. Effect of particle size on in vitro ruminant digestibility.

L.O. Pordesimo et al. / Bioresource Technology 96 (2005) 47–53 51

ber or lint content had lower digestibilities. This result

was highly significant ðP < 0:01Þ. It is also shown in

Fig. 1 that trash from the spindle-harvested cotton

generally had greater digestibilities than trash from the

stripper-harvested cotton ðP < 0:01Þ. For instance, trashfrom spindle-harvested cotton ground through a 0.5 mm

screen had an average digestibility of 58.8%, while that

from stripper-harvested cotton had an average digest-ibility of only 37.8% (both material obtained from West

Tennessee). This result is explained by the makeup of

the foreign material accumulated with the seed cotton

through each harvesting method. Stripper harvesting

includes a higher proportion of heavier and coarser

foreign materials (burrs and stems) that are inherently

less digestible because of their chemical composition.

The CGT from Texas, which was a mixture of all clea-nings, had a digestibility of 44.1%.

3.2. NaOH concentration

Pretreatment of the CGT with NaOH significantly

affected IVDMD ðP < 0:01Þ. It is evident from Fig. 2

that pretreatment of the CGT with either 4% or 6%

NaOH (w/w) improved IVDMD compared to no-

chemical treatment There was a significant type · che-

mical treatment interaction ðP < 0:01Þ. The general

trend in the figure indicates that treating the CGT with

6% NaOH seemed more effective in improving IVDMD

0102030405060708090

100

A B C D E F G H I J MCGT Type

IVD

MD

, % None4% NaOH6% NaOH

spindlestripper

Fig. 2. Effect of sodium hydroxide concentration on in vitro ruminant

digestibility.

than using 4% for all CGT types, with an average IV-

DMD of 70.5% compared to 60.6%. However, a missing

treatment combination prevented a statistical evaluation

of whether the decrease was linear with increasing NaOH

concentration nor could other mean comparisons

through orthogonal contrasts be undertaken. These

digestibility values obtained compare well to the 64%

achieved by submerging the CGT in a caustic solution inwhich a superoxide species, e.g. hydrogen peroxide, is

incorporated (Thomasson, 1990). It can also be seen in

Fig. 2 that high-lint trash types coming from both spin-

dle-picked and stripper-harvested cotton exhibited a

greater positive IVDMD response to NaOH pretreat-

ment than did the non-lint trash types.

3.3. Process temperature

According to chemical reaction kinetics, applying

heat increases chemical reaction rates. Based on this and

also because the starting ratio of CGT to water was keptclose to 1:2, the total time of mixing/chemical pretreat-

ment at elevated temperatures was reduced from the 4 h

used for most of the tests conducted. Preventing char-

ring of the CGT was another reason for this change. It

was presumed that the elevated process temperatures

would counterbalance the reduction in treatment time

thereby resulting in roughly the same IVDMDs achieved

at each temperature–time treatment. The process tem-perature–time treatments significantly affected IVDMD

ðP < 0:05Þ. As shown in Fig. 3, in the study conducted,

agitated NaOH treatment of the CGT at higher tem-

perature–shorter time resulted in a decrease in IVDMD.

However, a missing treatment combination prevented

an evaluation of whether the decrease was linear with

increasing temperature. A probable explanation for this

is the loss of water from the mixtures during the courseof processing. The average moisture content of the CGT

pretreated at �25 (ambient), 40 �C, and 50 �C was

30.51%, 28.34%, and 6.10% d.b., respectively. Water

serves as both the medium for pretreatment chemical

transport and a medium for heat transfer, so treatment

0102030405060708090

100

A B C D E F G H I J MCGT Type

IVD

MD

, % room temp40 °C50 °C

stripper spindle

Fig. 3. Effect of mixing temperature on the effectivity of 6% (w/w)

sodium hydroxide pretreatment.

Table 3

Pretreatment main effects on in vitro ruminant digestibility

Experiment no. Screen size (mm) Pretreatment Agitation time (h) Mean IVDMD (%)

NaOH concentration

(% w/w)

Process temperature

(�C)

1 0.5 0 – – 47.8± 1.15

1.0 0 – – 37.4± 1.15

2.0 0 – – 33.8± 1.15

2 2.0 4 �25 (room) 4 60.6a

2.0 6 �25 (room) 4 70.5± 2.10

3 2.0 6 �25 (room) 4 70.5± 4.18

2.0 6 40 3 65.2a

2.0 6 50 2 55.4± 4.18

a Standard errors could not be calculated due to missing treatment factor combinations.

52 L.O. Pordesimo et al. / Bioresource Technology 96 (2005) 47–53

effectiveness decreases as water evaporates during heat-

ing. This result shows that increased process tempera-

ture did not make up for the loss of water and maybeeven processing time when other process variables were

kept constant. Room temperature samples produced an

IVDMD of 70.5% while samples processed at 50 �Cproduced an IVDMD of only 55.4%. It is likely that in

order for elevated temperatures to effect improved

digestibility, additional water must be continuously ad-

ded to compensate for evaporation during the course of

the chemical pretreatment process.

4. Conclusions

Experiment 1 clearly indicated that size reduction was

effective in enhancing IVDMD of CGT. The effect of

grinding was greatest in the non-lint types. Under

intermittent agitation, a concentration of 6% NaOH

(w/w) was shown in Experiment 2 to be more effective in

improving CGT digestibility than was a 4% concentra-

tion. Lastly, Experiment 3 showed that undertakingagitated chemical pretreatment at elevated temperatures

over a short time did not enhance IVDMD because

water, the medium for chemical transport and heat

transfer, was evaporated. The average IVDMD is shown

in Table 3 for all experiments. For the range of experi-

mental treatment factors studied, the most effective

pretreatment of CGT was to grind the material through

a 0.5 mm screen and then treat the CGT with 6% NaOH(w/w) under agitation at room temperature for 4 h. This

certainly may not be the most efficient method when

considering cost and practicality under commercial

conditions. For instance, the extra power and/or time

required to grind material through a 0.5 mm screen may

not be worth the degree of IVDMD enhancement when

compared to IVDMD levels for CGT ground through a

2.0 mm screen. An optimal target IVDMD needs to bedetermined considering these factors.

This research study has only addressed the problem

of CGT digestibility. The presence of any residual pes-

ticides in the CGT was not investigated. The latter

represents a second hurdle that must be addressed be-

fore the use of CGT as a ruminant feed ingredient can beeffectively moved forward.

Acknowledgements

Support for this research from the Cotton Founda-

tion, the Southern Cotton Ginners Association, and the

Tennessee Agricultural Experiment Station (ProjectNumber TEN 230) is gratefully appreciated. Thanks are

also due to Dr. Arnold Saxton of The University of

Tennessee, Department of Animal Science for assistance

in the statistical analysis.

References

Agrawal, B.D., Verma, L.R., Craig, W.M., 1989. Chemical treatment

of lignocellulosic material for use as ruminant feed. Trans. ASAE

32 (6), 2097–2102.

ASTM Standard D 2495, 1987. Standard test method for moisture in

cotton by oven-drying. Textiles-Yarns, Fabrics, and General Test

Methods 7.01, 688–693.

Buser, M.D., 1999. Update on the extrusion of cottonseed and cotton

gin waste mixtures, In: Proc. Beltwide Cotton Conf., Orlando, FL.

3–7 Jan. 1999. Natl. Cotton Counc. Am., Memphis, TN, pp. 1468–

1472.

Conner, M.C., Richardson, C.R., 1987. Utilization of cotton plant

residues by ruminants. J. Anim. Sci. 65, 1131–1138.

Han, Y.W., Catalano, E.A., Ceigler, A., 1983. Treatments to improve

the digestibility of crop residues. In: Soltes, E.J. (Ed.), Wood and

Agricultural Residues: Research on Use for Feed, Fuels, and

Chemicals. Academic Press, NY, pp. 217–238.

Klopfenstein, T.J., 1978. Chemical treatment of crop residues. J.

Anim. Sci. 46 (3), 841–848.

Klopfenstein, T.J., 1980. Increasing the nutrient value of crop residues

by chemical treatment. In: Huber, J.T. (Ed.), Upgrading Residues

and By-Products for Animals. CRC Press, Boca Raton, FL, pp.

39–60.

Lalor, W.F., Jones, J.K., Slater, G.A., 1975. Cotton gin trash as a

ruminant feed. The Cotton Gin and Oil Mill Press 76 (6), 28–29.

McDougall, E.I., 1948. Studies of ruminant saliva.I. The composition

and output of sheep’s saliva. Biochem. J. 43, 99–109.

L.O. Pordesimo et al. / Bioresource Technology 96 (2005) 47–53 53

Moore, J.E., Mott, G.O., Dunham, D.G., Omer, R.W., 1972. Large

capacity in vitro OM digestion procedure. J. Anim. Sci. 35 (1), 232.

SAS Institute, Inc., 1999. The SAS System Online Doc, Version 8. CD-

ROM. Cary, North Carolina: SAS Institute.

Thomasson, J.A., 1990. A review of cotton gin trash disposal and

utilization. In: Proc. Beltwide Cotton Prod. Res. Conf., Las Vegas,

NV. 9–14 January 1990. Natl. Cotton Counc. Am., Memphis, TN,

pp. 689–705.

Thomasson, J.A., Anthony, W.S., Williford, J.R., Johnson, W.H.,

Gregory, S.R., Calhoun, M.C., Stewart, R.L., 1998. Processing

cottonseed and gin waste together to produce a livestock feed. In:

Proc. Beltwide Cotton Conf., San Diego, CA. 5–9 January 1998.

Natl. Cotton Counc. Am., Memphis, TN, pp. 1695–1698.

Tilley, J.M.A., Terry, R.A., 1963. A two-stage technique for the in

vitro digestion of forage crops. Brit. Grassland Soc. 18 (2), 104–

111.