TEXAS AGRICULTURAL EXPERIMENT STATION
R. D. LEWIS. Duector, Colleje Sicnion. Texas

 

Kullelin 7.2a

The EFFect of C|eanin3 on tHe Grade,
Stap|e and Price 0F Cotton

MARY ANNA GRIMES
Department 0f Rural Home Research

/'  A FY
.  , h rs‘, _
p“ g, M. CSJLEEGE OF TEXAS

GIBB GILCHRIST. Chcruellor

 

[Blank Page in Original Bulletin]

    
    
 
  
   
  
    
   
   
  
   
   
    
 
   
  

Digest

f The removal 0f all foreign matter from ginned cotton should
3' ' - the grade, qualifying it for a higher price. The extent
. the changes produced by cleaning with a Shirley analyzer
as determined for 132 cottons representing 31 varieties and
rains", 5 locations and 4 methods of harvesting. Two classers
f?» ed each cotton before and after cleaning, using the usual
ssifications for grade and staple. Price difierences were cal-
g lated at prevailing prices for normal ginned cotton.

l. Cleaning improved the average grade 2 grades according to
, e classer and 4 grades according to the other classer. Cleaning
Lid no significant elfect on staple.

- Changes in price due to cleaning, calculated for grades and
r 'ces of normal ginned cotton, ranged from a loss of $21 to
gain of $49 with average gains of $6 and $10 per bale for the
43,0 classers.

_i The classers dilfered an average of 0.3 grade for uncleaned
j d 2.2 grades for cleaned samples. Dilferences between classers
ere 2 or more grades for 34 uncleaned and for 82 cleaned
 l ples. In staple they dilfered approximately 3/32 inch for
écleaned and 2/32 inch for cleaned samples. These differences

f standards for grades and that estimates of length are not
pays reliable.

; Prices received by growers are based on government standards
'1' grade. The standards used by individual firms and classers
gquently are modifications of government standards. This prac-
e of using different standards, plus inevitable human errors,
 y add up to material losses to the growers. Government grade
 the chief determinant of price, therefore, it is tremendously
portant that each and every sample is graded as close to
j" ual government grade as is humanly possible. Frequent use
_‘_' the physical measurements now available should help reduce
e differences between classers.

; From this study, it is concluded that, at present classifications
jdprices, it probably would not. be profitable to remove all
“reign matter from cotton containing less than approximately
{percent

 -750-5M-L1 so
i

I: i

f g evidence that classers do not adhere to the same degree .

c 0 N T E N T s  
Page

Digest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 5
Preliminary Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 5
Methods of Study ................................................ .. s 
Waste Removed from Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 

Report for Classer A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7 i"

Uncleaned Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7 I

Cleaned Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . L . . . .. 8 

Elfect of Cleaning on Grade and Staple . . . . . . . . . . . . . . . . . . . . . . . .. 9 3

Etfect of Cleaning on Price . . . . . . . . . . . . . . . . . . . . . . . .  . . . . . . . . .. 11 ._
Elfect of Method of Harvesting and Location on Improvement by I
Cleaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 12"?

Report for Classer B . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15 .

Uncleaned Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 151

Cleaned Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 16;

Effect of Cleaning on Grade and Staple . . . . . . . . . . . . . . . . . . . . . . . . . 17 

Elfect of Cleaning on Price . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 . . . . .. 181i
Elfect of Method of Harvesting and Location on Improvement by ' i’
Cleaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 19 i"

Fibrograph Measurements of Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 212':

Comparison of Classifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 23

Grade and Staple of Uncleaned Cotton . . . . . . . . . . . . . . . . . . . . . . . . .. 23f

Grade and Staple of Cleaned Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . .. t

Changes in Grade and Staple Due to Cleaning . . . . . . . . . . . . . . . . . .. 28f

Price . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  .  31p.

Other Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 

Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 35f

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 381

ULLETIN 723 JUNE 1950

   
  
   
  
 
 
    
  
 
   
   
   

 he Effect of Cleaning on the Gracie,
Staple and Price of Cotton

MARY ANNA GRIMES, Professor
Department of Rural Home Research

 LL COMMERCIAL RAW COTTON contains some foreign
 matter. This foreign matter largely determines the grade,
fwhich, with the length, determines the price. The kind and quan-
pity of foreign matter are, therefore, of great importance not
{only to the grower and the cotton buyer but to all those who use
Lproducts made from cotton.

I The grower is constantly urged to harvest his cotton as clean
as possible so that he may receive a high price per pound.

7 Normally, hand-picked cotton contains the least foreign matter
qfollowed by machine-picked, snapped and machine-stripped. The
greatest difference in amounts of foreign matter is usually
found between hand-picked and each of the other methods.

. More and better mechanical harvesting is largely responsible
for the development of superior cleaning equipment, which,
with improvements in ginning equipment, are turning out clean-
er cotton. '

; The questions arise as to what extent additional time, labor
and machinery are justified by the higher price paid for the
"fcleaner cotton, and to What extent our present method of class-
iing cotton is satisfactory for clean cottons. This study was
i undertaken with the hope of answering some of these questions.

Preliminary Work

It was found in a preliminary study‘ that the removal, with
12a Shirley analyzerz, of all foreign matter from 60 cottons raised
the grade, according to a board of cotton examiners, an average
;of 4.2 grades. Improvement in grade due to cleaning resulted in
 changes in price from a loss of approximately $14 to a gain of
$24 per bale, a range of $38. The average- gain was $7 per bale.

 

, lThe Efi-‘ect of Foreign Matter on the Grade, Staple and Price of Cotton,
A by Mary Anna Grimes. 1945. Texas Agricultural Experiment Station
' Progress Report 954.

1 2The Shirley Analyzer, by George Pfeiifenberger. Textile Research, 1944,
. Vol. XIV, No. 2, pages 50-54.

   
   
 
   
  
   
 
 
 
 
 
 
 
   
 
    
  
   

6 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

When classed by a second classer, the averages were approxi- ,;
mately the same, although there were- frequently wide differ- '_
ences between the classers in their estimation of the effect of i
cleaning 0n the grade and staple of individual cottons. l

Methods of Study

To check the conclusions drawn from the preliminary work, f
the study was continued with 132 additional cottons which were .
cleaned with the Shirley analyzer and classed both in the un-' ,
cleaned and cleaned states by 2 classers. a

The Shirley analyzer removes practically all foreign matter v
plus some short or broken fibers from the ginned cotton. This .
material is hereafter frequently called waste. a

The 132 cottons came from 5 locations in Brazos, Burleson 1
and Lubbock counties. They included 31 varieties and strains v
and 4 methods of harvesting. In some cases the samples repre- 3
sented different harvesting dates from the same location. Most _.
of the samples were ginned on commercial gins.

The classers were told that the cleaned samples were the same i
cottons as the uncleaned but with the foreign matter removed. s
The samples were so numbered that pairing of uncleaned with I
cleaned samples was impossible, except by chance. The classers ,
were asked to give separate grades for leaf, color and prepara- 1
tion, which are factors comprising grade, and a. composite grade 1
for each sample. In most cases the preparation for the uncleaned
samples was said by the classers to be “normal” and no grade I
given for preparation. Since the preparation of the cleaned ;
samples was not “normal,” no grade was given for preparation. o-
For bo-th the uncleaned and cleaned, 3 grades were given for
each sample—leaf, color and the composite grade. The classer’s ‘
estimate of staple was given for each sample. ‘

As is routine procedure when cotton classing is requested, i
each classer was sent duplicate samples of uncleaned cottons
to see how closely each checked with his own classing. ‘

Prices used for both uncleaned and cleaned cotton are those 5i
for spot cotton for delivery March 1, 1948. Staple premiums and
discounts are averages for lengths quoted in New Orleans, Gal-
veston, Houston, Dallas, Memphis and Little Rock. Differences 1
for grade are the averages of the first 5 of these 6 markets.

Since in several cases the classers differed widely, the clas-si-j‘
fication given by each classer is shown. The report is, therefore;
divided into three sections, one for each classification and a
third for a discussion of the results. ?

Waste Removed from Cotton

The Shirley analyzer removes practically all visible foreign.’
matter and small amount of short or broken fibers. There are

  
  
    
  
   
  
    
    
  
   
 
 
 
 
  
   
  
   
 
  

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 7

only occasional bits of foreign matter visible after the lint
‘s twice passed through this cleaning device. The cleaned
omples, for which the classer was asked to assign grade and
.; ple, were almost completely devoid of foreign matter.

, The Waste removed by the Shirley analyzer from the 132
 ttons ranged from 3.8 to 23.3 percent by weight. The lowest
verage waste was in the hand-picked cotton followed in order
y machine-picked, hand-snapped and machine-stripped (Table
_). For the 59 hand-picked cottons, the range in waste was
i om 3.8 to 9.8 percent. The waste in the 27 hand-snapped
'nged from 5.1 to 16.0 percent. The 37 machine-stripped
nged from 4.2 to 23.2 percent. For the 9 machine-picked
ttons, the range in waste was the narrowest of the 4 methods,
,.1 to 9.7 percent.

A Report for Classer A
 ncleanedl Cotton

 The effects of cleaning on the grade, staple and price of
5.. cottons, based on the classifications of classer A, are given
u TQbIQS 1 t0 

l" Grouped by method of harvesting, the averages of waste,
nd the grades and staples for the uncleaned and cleaned
mples are shown in Table 1.

 The average composite grade for the 132 uncleaned samples
lfas grade 6.6, or approximately LM-|-. There was a range
' om grade 5(M) to grade 8(SGO). The hand-picked was
‘ven the highest average composite grade, 6.2, followed by
fpped, 6.7, machine-picked, 6.9, and stripped, 7.1 (Table 1).
g ere were differences of 0.5, 0.7 and 0.9 grades between
nd-picked and each of the other 3 methods of harvesting.

l The hand-picked cottons with the least waste, 5.9 percent,
mere classed 6.2 (SLM), and the machine-stripped with the

 

  The average waste, grade and staple of uncleaned and cleaned cotton*

Uncleaned Cleaned
Grade Grade

No. i i Staple, ~—,——— ——— Staple,

of Waste, Com- 32nd ' Com 32nd

_ cottons % Leaf Color posite inch Leaf Color posite inch
. 59 5.9 6.0 6.0 6.2(SLM) 27.7 3.1 4.6 4.2(SM ) 27.9
_ 9 3.6 6.9 6.3 6.9(LM ) 23.1 3.3 5.2 4.6( M+) 23.1
pped . . . . . .. 27 9.4 7.0 6.3 6.7(LM+) 21.0 3.3 4.9 4.5( M+) 27.4
"lripped..... 37 10.5 7.1 6.7 7.l(LM ) 25.0 3.5 5.5 5.0( M ) 26.7
. . _ . . . . . . . . . . . . . . . . . .. 8.1 6.6 i 6.3 6.6(LM+) 26.3 3.3 5.0 4.5( M+) 27.4

 

   
    

n grades are designated as follows:
ddling Fair (MF)

__~~ ct Good Middling (SGM)

_ d Middling (GM)

~ ct Middling (SM)

‘idling (M)

Strict Low Middling (SLM)
Low Middling (LM)
Strict Good Ordinary (SGO)
Good Ordinary (GO)

EDGJQG‘)
6.:-

s BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

most Waste, 10.5 percent, were classed 7.1 (LM), a difference o

of nearly 1 grade. The machine-picked and hand-snapped cot-
tons With little difference in Waste, 0.8 percent, dlfiered only
0.2 grade. The average percentages of Waste correspond rather
closely With the average composite grades of the uncleaned
cottons.

Leaf grade is based on foreign matter, therefore, it Would
be expected that the average grades for leaf of the uncleaned
cotton Would correspond closely With the average percentage of
Waste. This was true. There was approximately 1 grade difference
between the hand-picked with 5.9 percent Waste and the ma-

chine-picked With 8.6 percent. There Was a difference of only if
0.1 grade for leaf between machine-picked and hand-snapped p
and approximately 1 percent difference in Waste. There Was _.
also a difference of 0.1 grade and 1 percent Waste between a

hand-snapped and machine-stripped.
For individual cottons, the Wastes frequently do not cor-

respond With the leaf grades. The Waste of 1 cotton with a 1
leaf grade of 4 exceeded in Waste 6 of 18 cottons With a grade 1
of 5, exceeded 4 of grade 6 and 1 of leaf grade 7. Cottons 1
which contained 4.0 to 4.5 percent Waste Were found in leaf i.

grades 5, 6 and 7. Wastes of 10 and 12 percent Were found in

each of leaf grades 6, 7 and 8. These discrepancies indicate that i» .
grade for leaf is frequently not a reliable indication of the l
amount of Waste in a given cotton. The explanation in some '

cases may lie in the type of foreign matter. Cotton With fine,

heavy, inconspicuous Waste, such as sand, might be given a -
higher grade, but Would have a higher percentage of Waste by :_
Weight than a cotton With leaf trash Which Would occupy a 

larger area, but Weigh less.

The averages of grades for color Were, in general, lowered 
as the percentage of Waste increased, although there were 5

many exceptions for individual cottons.

The grades for leaf and color of the uncleaned cottons Were i
identical for 82 of the 132 cottons. The Waste for 1 cotton, l,
Which Was given a grade of 4 for color, Was greater than all 3
but 3 of the 19 cottons graded 5, and greater than all but 12 ;
of the 53 graded 6 for color. In many cases the percentage i
of Waste shows no apparent reason for a sample being given ;
a lower grade for leaf than for color. For example, 7 of the 9 '1
samples with grades 5 for color were graded 6 for leaf, yet 

Were among the cottons With the least Waste.
Cleaned Cotton

\

It might be expected, since the cleaned samples Were prac- f;
tically devoid of foreign matter, that the grade for leaf would ,
be high and the grade for color more variable, depending on
the presence of dullness, stains, tinges or other color defects. g

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 9

 The extent t0 which this assumption is true for averages for
 each method of harvesting is shown in Table 1.

 The composite grades for the cleaned cottons differed a
i maximum of 0.8 grades between methods of harvesting, ap-
g proximately the same difference as for the uncleaned samples.
 Hand-picked was graded 4.2(SM) and hand-snapped 4.5. Ma-
_ chine-picked was classed 4.6(M-|-) and machine-stripped 5.0
 (M). There was a difference of 0.8 grade between the composite
 grades of the cottons with an average of 5.9 percent waste
I removeg and those from which 10.5 percent waste had been
> remove .

_l The average grade for leaf for all cottons was 3.3, approxi-
 mately SM+. The average grades for the 4 methods of harvest-
 ing were from 3.1 to 3.5, a difference of approximately 0.5
f grade.

A»; The grades for the color of all cottons were lower than those
 for leaf. The average color grade for the 132 cleaned cottons
 was 5.0. For the 4 methods of harvesting, the average grades
1 of the cleaned cottons were 4.6 for hand-picked, 4.9 for snapped,
4 5.2 for machine-picked and 5.5 for stripped. These are in the
.1; same order as for the leaf grades.

 

1 Effect of Cleaning on Grade and Staple

 The removal of waste from the 132 cottons raised the com-
, posite grade an average of 2.1 grades (Table 2). The cottons
 harvested by the 4 methods were improved approximately the
T same amount-2.0 grades for hand-picked, 2.1 grades for
 stripped, 2.3 grades for machine-picked and 2.2 for snapped.
 The hand-picked cottons with the least waste were improved
1* approximately the same amount as the stripped with the most
p Waste (Table 2).

~ The change in grade of individual cottons, how-ever, often
 bears little relationship to the amount of waste removed. For
i. example, the composite grades of 2 cottons with an average
i waste of 4.9 percent were lowered 0.5 grade by cleaning, but
‘ 1 cotton, with 5.6 percent waste, was improved 5 grades (Table

Table 2. The effect of cleaning on the grade, staple and price

Changes due to cleaning
Method N0. Grade
0f _ 0f Staple,
harvesting cottons Waste, Com- 32nd Pri¢e
% Leaf Color posite inch difierence
 Hand-picked . . . . . . . . .. 59 5.9 2.9 1.4 2.0 + .2 $+ 5.10

.: Machine-picked . . . . . .. 9 8.6 3.6 1.6 2.3 0 +ll.l7

‘_ Hand-snapped . . . . . . . .. 27 9.4 3.7 1.4 2.2 + .4 + 8.86

’ Machine-stripped . . . . .. 37 10.5 3.6 1.2 2.1 +1.7 +19.ll

.,Average . . . . . . . . . . . . . . . . . . . . .. 8.1 3.3 1.3 2.1 + .6 +l0.21

 

10 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION
3). Two samples with wastes of 7.9 percent were improved
only 0.5 grade, but 61, with approximately the same average
Wastes, were improved 2 and 2.5 grades. The changes in com-

posite grade for the 132 cottons ranged from a loss of 0.5

grade to an improvement of 5 grades. .

Table 3. Average waste for leaf, color and composite grades for each
change 1n grade
Leaf grade Color grade Composite grade
Changes No. of Waste, No. of Waste, No. of Waste,
in grade cottons % cottons % cottons %
—— . 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 4.9

—l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 6. 4 . . . . . . . . . . . . . . . . . . . .

0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 7.3 3 6.3

. 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 7.9

l . . . . . . . . . . . . . . . . . 3 5.2 52 8.4 l8 7. 7'

l . 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l 1 l . 0 12 8. 9

2 . . . . . . . . . . . . . . . . . 22 5. 9 29 8.2 46 7. 9

2. 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 5.7 15 7.8

3 . . . . . . . . . . . . . . . . . 51 7 . 7 16 8. 5 20 8 . 4

3. 5 . . . . . . . . . . . . . . . l 5. 1 . . . . . . . . . . . . . . . . . . . . 6 8.3

4 . . . . . . . . . . . . . . . . . 42 9. l 5 8. 7 7 ll .0

4. 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5 . . . . . . . . . . . . . . . . . 12 ll .4 . . . . . . . . . . . . . . . . . . . . 1 5.6

5.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6 . . . . . . . . . . . . . . . . . 1 5.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The leaf grades of the 132

cottons were improved by clean-

ing an average of 3.3 grades (Table 2). Hand-picked with the
least Waste, 5.9 percent, was improved the least--2.9 grades.
Machine-picked with 8.6 percent waste was improved 3.6 grade-s,
snapped with 9.4 percent waste 3.7 grades and stripped with
10.5 percent waste was improved 3.6 grades.

The leaf grade was improved by cleaning from 1 to 6 grades.
All but 4 of the 132 samples were improved from 2 to 5 grades.
Approximately equal wastes were contained in 5 cottons, 3 of
which were improved by cleaning 1 grade in leaf, 1 by 3.5
grades and 1 cotton by 6 grades (Table 3).

Cleaning improved the color an average of 1.3 grades (Table
2). Machine-picked cottons were improved 1.6 grades. The
cottons which were hand-picked and hand-snapped were each
improved 1.4 grades. The stripped cottons were improved 1.2
grades in color. There was a maximum difference of slightly
less than 0.5 grade in improvement in color grade between
methods of harvesting.

The color grades of 3 cottons were lowered 1 grade after
cleaning. The colo-r grades of 24 cottons were not changed by
cleaning. Improvements of 1 and 2 grades followed the cleaning
of 82 cottons. The greatest improvement was 4 grades for 5
cottons, with an average waste approximately the same as for
52 which were improved 1 grade and as for 16 which were im-
proved 3 grades. The amount o-f waste apparently had little

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 11

eflect upon the direction or extent of the change in grade for
color.

Cleaning did not afiect the average- length of the machine- '

picked cottons (Table 2). The average lengths of the hand-
picked and snapped cottons were slightly increased. The great-
est change was in the stripped with an average increase of
between 1/32 and 2/32 inch.

For individual cottons, however, the staple was changed by
removing the waste in all but 46 of the 132 cottons. Changes
were 1/32 inch for 25 samples, 2/32 for 33, 3/32 for 4, 4/32
for 20 and 5/32 inch for 4 samples. The average change in
length for the 132 cottons was an increase of approximately
1/64 inch. The amount of waste removed apparently did not
afiect the direction or extent of the change in staple of indi-
vidual cottons.

Effect of Cleaning on Price

It should be kept in mind that for the cleaned as well as the
uncleaned samples, the grades and prices used, the only ones
available, are those designed for ordinary ginned cotton con-
taining a normal amount of foreign matter.

No commercial equipment comparable with the laboratory de-
vice, the Shirley analyzer, is available, therefore, the cost of
cleaning cannot b-e taken into account.

The removal of waste from lint cotton affects in 3 ways
the price for which it qualifies. First, it reduces the number
of pounds available for sale from each 500-pound bale of un-
cleaned cotton. Second, it changes the grade. Third, cleaning
may either lengthen or shorten the staple.

The removal of waste for some of the 132 cottons would have
resulted in a loss in price despite the higher grade, even when
there was no loss in staple, since the weight of the cotton was
reduced sufficiently to oflset the gain won by the higher grade.
This was true for 17 cottons which lost from $0.15 to $8.84
per bale, an average loss of $4.71. The loss in weight due to
cleaning was from 20 to 44.5 pounds, an average of 31 pounds
for those which lost in price.

Twenty-three of the 59 hand-picked cottons lost by cleaning
from $0.15 to $20.98 per bale. Thirty-six cottons gained from
$0.23 to $38.42 per bale. The range in change of price was
$59.40. The average change was a gain of $5.10 per bale.

The average change in price due to the cleaning of the 9
machine-picked cottons was a gain of $11.17 per bale. Seven
cottons gained from $3.88 to $24.13. Two cottons lost $0.53
and $1.36 per bale. No machine-picked cotton showed a loss in
price due to cleaning where the staple was not changed in the
process.

12 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

Twenty of the 27 snapped cottons gained fro-m $3.74 to
$27.86 per bale. Seven 10st from $2.50 t0 $18.29. The average
change due to cleaning was a gain of $8.86 per bale.

Thirty-three of the 37 stripped cottons gained from $0.02
to $48.63 per bale. Four cottons, lost from $1.26 to $7.87. The
average change was a gain of $19.11 per bale. The range was
$56.50.

The price changes due to cleaning of the 132 cottons were
from a loss of $20.98 to a gain of $48.63, a range of $69.61.
There were 96 cottons which gained in value and 36 which lost
by cleaning. The average change in price due to cleaning was
a gain of $10.21 per bale.

Elfect of Method of Harvesting and Location on Improvement by Cleaning

If such information is available, classers are prone to place
much weight upon the region where the cotton was grown and
the method of harvesting, basing their judgment upon past ex-
perience with such cotton rather than on the quality of the
particular cotton being classed. They thus use their own stand-
ards instead of official government standards. This may result
in the quality of a cotton being classed either too high or too
low. For these reasons, the classers used in this experiment
were not told the region where the cottons were grown nor
the method of harvesting.

Four methods of harvesting were employed at College Sta-
tion. Approximately the same number of samples were obtained
in three of them. This made possible a comparison of improve-
ments following cleaning for each of the methods of harvest-
ing (Table 4).

Table 4. Etfect of cleaning on cotton from the same location harvested by
different methods

Changes due to cleaning
Location and No. Grade
method of Staple, Price
harvesting cottons Waste, Com- 32nd per
Leaf Color posite inch bale

College Station

Hand-picked . . . . . . .. 23 5.2 2.7 2.0 2 4 + .2 $ + 93

Machine-picked. . . .. 3 7.7 3.7 1.3 2 0 —.3 +10 22

Hand-snapped....... 20 9.4 3.8 1.5 2 4 + .2 + 9 41

Machine-stripped. . . . 22 11.3 4.0 1.6 2 4 +1.2 +20 51
Chance Plantation

Hand-picked . . . . . . .. 7 5.9 2.9 1.4 1.8 -— .3 -—3.91

Machine-picked..... 6 _9.0 3.5 1.7 2.4 + .2 +11.64
Snook

Hand-picked . . . . . . .. 11 4.8 2.4 1 1 1.8 —l.0 -— 4.70

Hand-snapped....... 2 11.1 3.0 5 .8 — .5 + .75
Lubbock

Hand-snapped . . . . . .. 5 8.8 3.6 1.2 2 1 +1.2 + 9.92

Machine-stripped. . .. 15 9.4 2.9 .8 1 7 +2.5 +17.05

 

 

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 13

The average Waste ranged from 5.2 percent for hand-picked
to 11.3 percent for machine-stripped cottons. The improvement
in composite grade Was identical for 3 of these methods, 2.4
grades, and for the fourth, machine-picked, 2.0 grades. In only
one case, the machine-stripped, did the classer find a signifi-
cant change- in staple, a gain of 1/32 inch. -

The average changes in price due to cleaning ranged from
a gain of $3.93 for the hand-picked, with 5.2 percent Waste
removed, to a gain of $20.51 per bale fo-r the machine-stripped
with 11.3 percent waste removed.

Samples of Deltapine cotton from 13 bales Were obtained
from the Chance plantation near College Station. Seven were
hand-picked and 6 machine-picked. The 2 methods of harvest-
ing were carried on simultaneously. These samples offer the
best comparison of these 2 methods of harvesting, as the
variety, location and dates of harvesting were the same. The
waste of these 13 samples and the changes in grade, staple
and price due to cleaning are given in Table 4.

The 7 hand-picked samples contained an average of 5.9 per-
cent Waste, and the 6 machine-picked, 9.0 percent Waste. The

 hand-picked uncleaned samples were given an average com- '
j? posite grade of 5.9 and the machine-picked a composite grade
. of 6.8, a difference of approximately 1 grade due to method
I» of harvesting. Both the hand-picked and machine-picked samples
> were given an average staple of 29/32 inch. Cleaning had no
 significant effect on the staple.

The average composite grade for the 7 hand-picked cottons

 was 4.1 after cleaning, an improvement of 1.8 grades over the
 uncleaned samples. The leaf grade was improved by cleaning
,' 2.9 grades and color 1.4 grades.

The 6 machine-picked cottons were given an average com-

 posite grade of 6.8 before cleaning and 4.4 after the removal
, of 9 percent Waste, an improvement of 2.4 grades.

The average leaf grades of the 6 machine-picked samples wer

‘ 7.0 before cleaning and 3.5 grades after cleaning, an improve-
 ment of 3.5 grades. The average color grades were 6.7 before
cleaning, and 5.0 after cleaning, an improvement of 1.7 grades.
Q The machine-picked with 3 percent more waste were improved
 approximately 0.5 grade more in leaf, color and composite
a grade than the hand-picked.

The 7 hand-picked samples lost an average of $3.91 per bale

 by the removal of 5.9 percent waste. The 6 machine-picked
A samples gamed an average of $11.64 by the removal of 9 per-
 cent waste.

Although 2 samples are not enough for reliable conclusions,

. comparisons of the samples from Snook (Table 4) show that

 

14 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

the 2 snapped samples were improved 1 grade less than the 11
hand-picked samples. The hand-picked lost 1/32 inch by clean-
ing. The removal of 4.8 percent waste from the hand-picked
decreased the price $4.70 per bale. Removing 11.1 percent
waste from the snapped increased the price only $0.75 per bale.

The snapped and stripped samples from Lubbock contained
approximately the same percentage of waste. Cleaning raised
the composite grade of snapped 2.1 grades and increased the
staple 1/32 inch. The increase in price per bale was $9.92. The
stripped was improved 1.7 grades, gained more than 2/32 inch
in staple and $17 .05 per bale.

The effect of cleaning of cottons harvested by the same
methods at different locations is shown in Table 5.

Table 5. Effect of cleaning on cotton harvested by the same methods at
diiferent locations

Changes due to cleaning
Grade
MGthOd 0f N0. Staple, Price
harvesting of Waste, Com- 32nd per
and location cottons % Leaf Color posite inch bale

Hand-picked

College Station . . . . .. 23 5.2 2.7 2.0 2.4 + .2 $ + 3.93

ChancePlantation... 7 5.9 2.9 1.4 1.7 —— .3 — 3.91

Snook . . . . . . . . . . . . .. 11 4.8 2.4 1.1 1.8 —l.0 —4.70

Main Station Farm... 18 7.5 3.7 .9 1.9 + .9 4-16.08
Machine-picked .

College Station . . . . .. 3 7.7 3.7 1.3 2.0 ——— .3 +l0.22

Chance Plantation. . . 6 .0 .5 1.7 2. + .2 +ll.64
Hand-snapped

College Station . . . . .. 20 9.4 .8 1.5 2.4 + .2 + 9.41

Snook . . . . . . . . . . . . .. 2 11.1 3.0 .5 .8 — .5 +0.75

Lubbock . . . . . . . . . . .. 5 8.8 .6 1.2 2. +1. + 9.92
Machine-stripped

College Station . . . . .. 22 ll .3 3.9 1 .6 2.4 +1 .2 +20.5l

Lubbock . . . . . . . . . . .. 15 9.4 2.9 .8 1.7 +2.5 +l7.05

The 18 hand-picked samples from the Main Station Farm
were of the same variety but represented difierent picking
dates. An interval of 36 days elapsed between the first and
last pickings. No sample was exposed to more weathering than
another, as all cotton open on each picking date was picked.
Differences in the amount of waste between picking dates were
no greater than between samples harvested on the same date.
There appeared to be no» significant differences in grade and
staple, either before or after cleaning,- which could be attributed
to the date- of picking.

In general, the cottons with the most waste removed by clean-
ing were improved the most in leaf grade and in price per
bale. This increase was, however, not always due to a greater
improvement in grade over those with less waste but to an
increase in staple after cleaning. .

 

7

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON l5

Three of the locations Where 59 cottons were hand-picked
represent single farms or plantations and are called in this
study, College Station, Main Station Farm and the Chance
plantation. The fourth location, Snook, includes 4 varieties of
cotton from 10 farms.

There was only 0.1 grade difference between the improve-
ment in the composite grade of the samples from Snook with
the least waste, 4.8 percent, and those from the Main Station
Farm with the most waste, 7.5 percent. The cottons from Col-
lege Station contained 5.2 percent Waste and were improved
the most by cleaning.

There were no great differences in percentage of waste and
in improvement in composite grade following cleaning between
the 2 locations where cotton was machine-picked.

The cottons snapped at College Station and Lubbock con-
tained approximately the same percentage of waste. They were
improved by cleaning approximately the same amount both in
grade and price. However, those at Lubbock were 1/32 inch
longer after cleaning. The 2 samples from Snook contained
more waste than those from College Station or Lubbock, and
welre improved by cleaning less in grade and by only $0.75 per
ba e.

The 22 machine-stripped cottons from College Station con-
tained more waste and were improved more in grade by clean-
ing than the 15 stripped at Lubbock. The College Station sam-
‘ples were 1/32 inch longer and the Lubbock samples more than
2/32 inch longer after cleaning. The increases in price were
$20.51 for the College Station stripped and $17.05 for the Lub-
bock stripped cottons.

Difierences in the percentage of waste do not always explain
the differences in grade, staple and price either before or
after cleaning.

Report for Classer B
Uncleaned Cotton

Grouped by method of harvesting, the average wastes, and
the average grades and staples assigned by classer B to the
uncleaned and cleaned cottons, are given in Table 6.

The hand-picked with the least waste received the highest
average grade for leaf, color and composite grade. There was
little difference between the grades assigned the uncleaned
samples harvested by the other 3 methods.

When the wastes are averaged for each leaf grade, it is
found that, in general, the cottons with the lowest average leaf
grade contained the most waste, however, there were many ex-
ceptions among individual samples. Seven of those cottons

 

16 ‘ BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

  
   
   
  
   
  
 
  
   
   
  

which were given a leaf grade of 6 contained less waste than "a
a number of those graded 4 for leaf. The one cotton which was “
given a grade of 3 for leaf contained more waste than any given T
a grade of 4, more than all but 3 of the 25 given a leaf grade 
of 5, and more than all but 12 of the 30 graded 6. One of .
the cottons with a leaf grade of 7 contained less waste than T
the 1 cotton graded 3.

The percentages of waste show no reason for some samples i
being given a lower grade for leaf than for color. For example, a
3 of the 8 graded 6 for leaf and 5 for color had less waste than _
the 1 cotton given a grade of 3 for leaf and 4 for color. The .
wastes of 2 of 17 cottons given a color grade of 5.5 were é
greater than all but 3 of the 19 graded 6.5, and greater than i
all but 9 of the 46 graded 7 and 7.5. a

The grades for color and leaf were identical for 77 of the 
132 cottons. 7'

The averages of grades for color were, in general, lowered 
as the percentage of the waste increased, although there were ~'
many exceptions for individual cottons. ‘

Cleaned Cotton

The leaf grades for the hand-picked and machine-picked f
samples were the same after cleaning, 1.4 grades. The snapped »
and stripped cottons were given the same leaf grade, 1.7 _'i
(Table 6). _ ’

The machine-picked samples were given a slightly higher
grade for color, 2.2, than the hand-picked, 2.5. Hand-snapped 
cottons were graded higher for color, 2.7, than machine-stripped, a.
3.2.

The average leaf grades were higher for each of the 4 Q
methods of harvesting than the average color grades. The
average leaf grade for the 132 cleaned cottons was 1.5 and ~
the avlerage color grade was 2.7, a difference of slightly over

gra e.

Table 6. The average waste, grade and staple of uncleaned and cleaned cot [i

Uncleaned Cleaned
Method Grade Grade
of No. ———— ————i— Staple, —-—

harvesting of Waste, Com- 32nd Com.

cottons % Leaf Color posite inch Leaf Color posite
Hand-picked . . . . . . . .. 59 5.9 5.6 5.8 5.8(SLM ) 30.1 1.4 2.5 2.l(SGM )
Machine-picked . . . . .. 9 8. 6 6.8 6.6 6. 7( LM +) 30.8 1.4 2.2 1.9(SGM )
Hand-snapped . . . . . .. 27 9.4 6.8 6.4 6.6( LM +) 29.4 1.7 2.7 2.3(SGM )
Machine-stripped..." 37 10.5 6.9 6.6 6.8( LM ) 28.8 1.7 3.2 2.8( GM )
Average . . . . . . . . . . . . . . . . . . . .. 8.1 6.3 6.2 s.3( LM+) 29.6 1.5 2.7 2.s( GM+)

 

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 17

  
 

The machine-picked cottons were given the highest com-

. posite grade, 1.9, followed in order by hand-picked, 2.1; snapped,
2.3; and stripped cottons, 2.8 grades. There was nearly 1 grade
difference between the machine-picked and machine-stripped
cottons. The average composite grade for the 132 cleaned cot-
tons was 2.3, approximately GM+.

 Effect of Cleaning on Grade and Staple

The composite grades of the 132 cottons were raised by
cleaning an average of 4.0 grades. The improvements follow-
ing cleaning ranged from 3.7 grades for the hand-picked to
4.8 grades for the machine-picked (Table 7).

 

Table 7. The elfect of cleaning on the grade, staple and price

 

Changes due to cleaning
Grade
Method No. Staple,
of , of Waste, Com- 32nd Price
harvesting cottons % Leaf Color posite inch dilference
Hand-picked . . . . . . . . .. 59 5.9 4.2 3.3 3.7 —.1 $ +1.00

Machine-picked . . . . . .. 9 8.6 5.4 4.4 4.8 0 +ll.4l

Hand-snapped . . . . . . . . . 27 9.4 5.1 3.7 4.3 —-.1 +9.08

Machine-stripped . . . . . . 37 10.5 5.2 3.4 4.0 +.2 +l0.74

Average . . . . . . . . . . . . . . . . . . . . .. s.1 l 4.8 l 3.5 4.0 o +6.09

a In leaf grade, the hand-picked cottons with the least waste
removed were improved the least of the 4 methods of harvest-
ing, 4.2 grades. Machine-picked samples contained slightly less
Waste than snapped or stripped samples, but were improved
more in leaf grade, 5.4 grades. The leaf grades of snapped and
stripped cottons were improved by cleaning approximately the
same amount, 5.1 and 5.2 grades, respectively.

The color grades of the cottons were improved less by the re-
moval of waste, 3.5, than were the leaf grades, 4.8.

The color grade of hand-picked cottons with the least waste
was improved approximately the same amount as the machine-
stripped with the most waste. Although machine-picked cottons
contained less waste, the color grade- was improved more by
cleaning than was the color grade of the snapped or the stripped
cottons.

The average wastes for leaf, color and composite grades for
each change in grade are given in Table 8.

The leaf grade was improved by cleaning from 2 to 6 grades.
The 2 samples which were improved 2 grades contained more
waste, 5.9 percent, than the 10 which were improved 3 grades,
4.4 percent.

The improvement in leaf grade due to cleaning was 4 grades
for 35 samples, 5 grades for 56 and 6 grades for 29 samples.

18 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

Table 8. Average waste for leaf, color and composite grades for each
change in grade

Leaf grade Color grade Composite grade
Changes No. of Waste, No. of Waste, N0. of Waste,
in grade cottons % cottons % cottons %
0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 7.2 . . . . . . . . . . . . . . . . . . . .

l . 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2 . . . . . . . . . . . . . . . . . . . 2 5 l6 7 4 3 5 . 1

2. 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 7 . l

3 . . . . . . . . . . . . . . . . . . . l0 4 4 49 8 0 16 6.4

3 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 7.8

4 . . . . . . . . . . . . . . . . . . . 35 6 3 47 8 2 20 8.0

4 . 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 8 . 4

5 . . . . . . . . . . . . . . . . . . . 56 9 . 2 l9 8 8 19 l0 . 0

5. 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 9. 1

6 . . . . . . . . . . . . . . . . . . . 29 9. 5 . . . . . . . . . . . . . . . . . . . . 1 10.4

The color grade of 1 cotton was improved by cleaning only 1 .4

grade. Ninety-six cottons were improved 3 and 4 grades. There
was a difference of only 1.6 percent Waste between the cotton
improved 1 grade for color and the 19 which were improved 5
grades.

Improvements in composite grade after cleaning were from
2 grades for 3 cottons with 5.1 percent Waste, to 6 grades for
1 cotton with 10.4 percent waste. In general, the improvement
in grade increased as the percentage of Waste removed in-
creased.

The average improvement in grade due to cleaning of the‘

132 cottons was 4.8 grades for leaf, 3.5 for color and 4.0 for
composite grade.

There was no significant change in staple due to cleaning for
cottons harvested by any of the 4 methods.

Elfect of Cleaning on Price

The removal of Waste from the 132 cottons resulted in an
average increase in price of $6.09 per bale (Table 7). Eighty-
two cottons gained and 5O lost in price per bale.

The 59 hand-picked samples, which had the least Waste, gained
the least, an average of $1.00 per bale. The change in price
was from a loss of $15.64 to a gain of $32, a range of $47.64.
Twenty-four samples gained from $0.01 to $32 and 35 samples
lost from $0.08 to $15.64 per bale.

The 9 machine-picked cottons gained an average of $11.41
per bale. Eight gained from $8.33 to $23.67, and 1 lost $8.36
per bale, a range of $32.03.

The average gain for the 27 snapped cottons was $9.08 per

bale. Six samples lost from $2.82 to $11.35 and 21 gained from ~

$2.49 to $33.78 per bale. There was a range in change of price
per bale of $45.13 for the snapped cottons.

    
 
    
   
   
   
  
   
  
  
   
  
 
   
  
  
  
  
  
    
   
 
    
   
   
 
 

i‘ - EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 19

; he average change in price due to cleaning of the 37 stripped
f» ns was a gain of $10.74 per bale. Gains ranged from $0.60
$34.83 per bale for 29 samples. Losses ranged from $0.23
$13.15 for 8 cottons. There was a range in the change in
'ce per bale of $47.98.

QThere were 15 samples of 1 variety from the same planta-
'n, 7 of which were hand-picked and 6 machine-picked. Each
the 7 hand-picked samples lost in price by the removal of
yste. The average loss was $2.48 per bale. Each of the 6
Mchine-picked samples gained in price. The gains were from
.33 to $15.96, an average gain of $12.63 per bale.

For these cottons the removal of 5.8 percent waste from the
nd-picked samples did not raise the grade and staple suf-
iently to compensate for the loss in weight. The removal of
"0 percent Waste from the machine-picked samples improved
~¢ grade to a degree which more than compensated for the loss
 weight.

 For this one variety and location, the removal of Waste from
1- hand-picked samples would have resulted in an average loss
 $2.48 per bale. In contrast, the removal of Waste from the
iachirie-picked samples would have resulted in a gain of $12.63
~ a e.

to ect 0f Method of Harvesting and Location on Improvement by Cleaning

 The improvement following cleaning was the same, 3.8 grades,
 the composite grade for 3 methods of harvesting at College
tation, but 0.6 grade more, 4.4 grades, for the snapped sam-
les (Table 9). The hand-picked and machine-picked samples
‘ere improved the same in color. The stripped were improved
pss and the snapped slightly more in color grade than were

 able 9. Effect of cleaning on cotton from the same location harvested by
. different methods

Changes due to cleaning
Grade
Location and No. staple, Price
method of of Waste, Com- 32nd per
harvesting cottons % Leaf Color posite inch bale

College Station

. Hand-picked . . . . . . .. 23 5.2 4.0 3.7 3.8 + .3 $ —0.l5
i Machine-picked..... 3 7.7 4.7 3.7 3.8 + .3 + 8.97
- Hand-snapped....... 20 9.4 5.2 3.9 4.4 + .1 +ll.37
? Machine-stripped. . . . 22 11.3 5.3 3.2 3.8 + .3 +15.0l
» hhnce Platation

'- Hand-picked . . . . . . . . 7 5.9 4.4 3 4 3.9 0 — 2.48
‘j Machine-picked. . . . . 6 .0 5.7 4 7 5.2 — 2 +12. 63
Bnook

1 Hand-picked . . . . . . .. l1 4.8 3.5 3.0 3.1 -—1.2 -— 7.60
i» Hand-snapped . . . . . . . 2 11 . 1 5. 5 2 . 5 3 . 2 —- .5 +21 72
lubbock

" Hand-snapped . . . . . .. 5 8.8 5.0 3.6 4.4 — .4 — 5.12
“V. Machine-stripped. . .. 15 9.4 4.9 3.5 4.3 + . l + 4.47

 

20 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

the hand-picked and machine-picked samples. The hand-picked
samples were improved the least in leaf, 4.0 grades, and the
stripped the most, 5.3 grades. Snapped samples were improved
approximately the same in leaf grade as were stripPQd samples.
Machine-picked samples were improved more’ in leaf than hand-
picked but less than snapped or stripped.

The maximum difference in change in price per bale due to
method of harvesting at College Station was $15.16. Cottons
which were hand-picked lost $0.15 per bale and those harvested
by the other 3 methods gained $8.97, $11.37 and $15.01, re-
spectively.

There was no significant difference in the change in staple

l following cleaning which was due to the method of harvesting

at College Station.

The machine-picked samples from the Chance plantation
were improved 1.3 grades more for leaf, color and composite
grade than were the hand-picked samples. There was no sig-
nificant difference in the staples. The hand-picked samples lost
$2.48 per bale following cleaning and the machine-picked gained
$12.63, a difference of $15.11 due to method of harvesting
(Table 9).

The 2 hand-snapped cottons from Snoo-k were improved 2
grades more in leaf than were the 11 hand-picked cottons. Hand-
picked samples were improved 0.5 grade more in color than
snapped samples. There was no significant difference in com-
posite grades between the 2 methods of harvesting. The picked
samples lost slightly more in length of fiber than did the snapped
samples. The hand-picked samples lost $7.60 per bale by cleaning
and the snapped gained $21.72, a difference of $29.32 between
the 2 methods of harvesting at Snook.

There was a difference of only 0.1 grade in leaf, color and
composite grades between the cottons which were snapped and
those which were stripped at Lubbock. There was a difference
of $9.59 in change in price following cleaning, as those which
were snapped lost $5.12 and those stripped gained $4.47 per
bale (Table 9).

Cottons harvested by the same method but at different loca-
tions differed from less than 0.5 to 1.5 composite grades in
their improvement following cleaning (Table 10).

The smallest differences between locations in improvement
in grade were for the stripped cottons from Colle-ge Station
and Lubbock. These differed 0.5 grade or less for leaf, color
and composite grade. There was a difference between the 2 lo-
cations of $10.54 per bale in the change in price.

The maximum differences between any 2 of the 4 locations
where hand-picking was used were 1.5 grades for leaf, 0.7

    
    
  
  
  
  
  
    
  
  
    
  
   
    
 
  
  
 
  
  
  
     
   
     
  

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 21

‘fade for color and 0.8 grade for the composite grade. At 3
f tions, the hand-picked cottons lost from $0.15 to $7.60
Q bale following cleaning. At the fourth location, the Main
tion Farm, greater waste was removed than at the other
locations and there Was a greater improvement in price due
* cleaning.

Cotton was machine-picked at 2 locations. There was a dif-
rence between the 2 locations of slightly over 1 grade and of
.66 per bale in the improvement following cleaning.

3' Among the 3 locations where cotton was snapped, there was
émaximum difference in improvement following cleaning of
 grade for color, 0.5 grades for leaf, 1.2 grades for composite
i de and $26.84 per bale (Table 10).

lible 10. Elfect of cleaning on cotton harvested by the same methods at
" dilferent locations .

 Changes due to cleaning
t Grade
Method of No. Staple, Price
harvesting of Waste, Com- 32nd per
and location cottons % Leaf Color posite inch bale
f‘ 1 d-picked
I} ollegeStation . . . . .. 23 5.2 4.0 3.7 3.8 —~— 3 $— 0.15
 hancePlantation... 7 5.9 4.4 3.4 3.9 0 — 2.48
F3 . . . . . . . . . . . . .. ll 4.8 3.5 3.0 3.1 —l.2 ——7.60
~ ainStation Farm... 18 7.5 5.0 3.1 3.6 + .1 + 9.08
‘ hine-piicked
a IlegeStation . . . . .. 3 7 7 4.7 3 7 3 8 —l— .3 + 8 97
f; hance Plantation. . . 6 .7 5 2 — .2 +12.63
‘A d-snapped
 llegeStation . . . . .. 20 9.4 5.2 3.9 4.4 + .1 +ll.37
60k . . . . . . . . . . . . .. 2 11.1 5.5 2.5 3.2 — .5 +2l.72
Vubbock . . . . . . . . . . .. 5 8.8 5.0 3.6 4.4 — .4 — 5.12
l hine-stripped
pollegeStation . . . . .. 22 11.3 5.3 3.2 3.8 + .3 +l5.0l
jbbock . . . . . . . . . . .. 15 9.4 4.9 3.5 4.3 + .1 + 4.47

iCotton was snapped at 3 locations, The leaf was improved
j- most at Snook, 5.5 grades, and the least at Lubbock, 5.0
iddes. The College Station cottons were improved most in
 and the Snook the least, a difference of 1.4 grades. The
gposite grades of the Lubbock and College Station samples
ire raised by the same amount, 4.4 grades, and Snook by
i’ grades. The maximum difference in change in price per
_e was $26.84, between Lubbock which lost $5.12 and Snook
ich gained $21.72.

Fibrograph Measurements of Length

j‘ there were wide differences between the classers, the
gths of the 132 samples both before and after cleaning were
f’ rmined with the Fibrograph to see if this method would
eal any appreciable change in length due to cleaning. It
A‘; found that the average change was not statistically sig-

22 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

Table 11. Classers’ staple and Fibrograph upper half mean before and
after cleaning

 

Classers’ mean length and upper half mean. inches Uncleaned t Cleaned
Classer A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .84 :1: .006 .86 d: .004
Classer B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93 d; .005 .93 d; .004
Fibrograph upper half mean . . . . . . . . . . . . . . . . . . . . . . . .. .91 d: .007 .91 d: .007

   

nificant, but that there appeared to be a slight tendency for f

long-fibered cotton to be slightly shorter and short-fibered

cotton to be slightly longer after cleaning (Table 11 and Figure t

1). Presumably, some of the longest fibers might have been

broken and some of the shortest fibers removed during the 

cleaning process.

It was found that the average upper half mean for the un- "
cleaned samples was .91 inch. Conversion of staples for classers ~
A and B to decimals gives .84 inch for A and .93 for B. For ‘
the cleaned samples, the upper half mean is also .91 inch, the .
staple for A is .86 and for B .93 inch. The measurements made ‘
with the Fibrograph gave a slightly higher correlation with 
the staples assigned by classer B, .72 for uncleaned "and .69 *1
for cleaned, than for those with classer A, .56 for the uncleaned 5

and .58 for the cleaned samples.

The staple as given for classer B for both the uncleaned ‘-
and cleaned samples agreed closely with the measurements "
given by the Fibrograph. It appears that classer A routinely 1'
called the staple short, but his results for the uncleaned and 7'
cleaned samples are consistent. It appears both by the Fibro- =
graph and the 2 classers that the cleaning had no effect on the =

average length of the fiber.

I a4 1.06 - /<>
o

E 33 L03 "

u
w
S
5
o

in
m
co

in
u
Q

.906

.875

.844

w
o
FIBROGRAPH IN INCHE
in
5

25 .78l

24 .750

+5 +4 +3 +2 _ +| o -| -2 —3
CHANGE lgN STAPLE IN 32ND INCH

Figure 1. Relation between length and change in staple when cleaned.

   
  
  
   

  
 
  
  
  
   
   
  
 

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 23

i Comparison ‘of Classifications
'21- and Staple of Uncleaned Cotton

}Of the 5 cottons sent in duplicate (10 samples) to one classer,
' same grade was assigned for 3 cottons, 1 of which also Was
en the same staple. Duplicate samples differed 0.5 grade
Q. 1 cotton and 2 grades for the fifth cotton. In staple, the
vllicate samples differed 1/32 inch for 1 cotton, 1/16 inch for
and 1/8 inch for 1 cotton. The last named cotton was said
be slightly irregular, which may account for such a Wide
erence.

‘ Of the 10 cottons sent in duplicate (20 samples) to the other
sser, the same grade Was assigned to duplicates of 3 cot-
4s, 2 of which were given the same staple. For the remaining
fecottons, duplicate samples differed 0.5 grade for 2, 1 grade
r 3 and 1.5 grades for 2 cottons. Of the 10 cottons, duplicate
 ples of 5 were given the same staple, duplicates of 1 cotton
gfered 1/32 inch, 2 differed 1/16 inch and 2 differed 1/8 inch.
ve of these 10 cottons were said to be slightly irregular, but
“ey included 3 of the 5 for which duplicate samples were given
 same staple and included only 1 of those for which duplicate
' ples differed 1/8 inch.

Differences between the grades and staples of duplicate sam-
 resulted in differences in price per bale ranging from $2.30
$39.25 for one classer and from $5.30 to $31.50 for the other
418891‘.

 It is believed that it would be difficult to find classers superior
I judgment and experience to those who classed the cottons
~.- in this study, but it is seen that they were frequently un-
le to check with themselves. It cannot be said that these
-' classers were asked to class duplicates of unfamiliar
tton, since they were normal, ginned Texas cottons not yet
ned by the Shirley analyzer.

   
 
  
  
  
   
   
 
    
   
    
  
  

ble 12. Differences between classers by grade giving number of cottons
s; for each grade difference

Number of cottons
Uncleaned Cleaned
Grade Leaf Color Composite Leaf (‘olor Composite
diflerence grade grade grade grade grade grade
. . . . . . . . . . . . . . . . . 68 74 55 5 4 3

. . . . . . . . . . . . . . .. 1 1 43  6

. . . . . . . . . . . . . . . . . 60 52 28 45 26 ll

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 4  2 30

. . . . . . . . . . . . . . . . . 2 5 1 65 51 21

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 1  27

. . . . . . . . . . . . . . . .. l  15 30 12

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . _ . . . . . . . . . . . . . . . . . . 21

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2 19 

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l
,.~ ber differing by 2
ore grades . . . . . . 3 5 2 82 100 82

24 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

The magnitude of the difierences in grades assigned by the 
2 classers are shown in Table 12. The 2 classers agreed closely 
on the average grades of the 132 uncleaned cottons but fre- a
quently differed _widely for individual cottons. There was a I
range in grade from 5 to 8 grades for classer A and from 4  -
to 8.5 grades for classer B. The classers differed only 0.3 grade .
for the average composite grade of the 132 uncleaned samples 
but agreed on the composite grade of only 55. They differed '9
from 1 to 2.5 grades for the composite grades of 34 cottons. l

The classers differed 0.3 grade for the average leaf grade. j
They gave the same leaf grades to 68 uncleaned samples. They is
differed 1 grade for 60, 2 grades for 2 and 3 grades for 1 sample. ..

The difference between classers for the average color grade a
was only 0.1 grade, but 57 samples differed 1 and 2 grades. j
The color grades for 74 samples were identical. ‘

Classer A gave a difference of 0.9 between the composite 
grades of the cottons with an average of 5.9 percent waste;
and those with 10.5 percent waste. For the same cottons, classer Y’
B gave a difference of 1 grade. In general, the highest average ~
grades were assigned to samples which contained the least ‘j
waste. For both classers, the cottons with intermediate wastes, ‘
8.6 and 9.4 percent, the grades were also intermediate, but not (
in the same order. Classer A found the cottons with 9.4 percent i;
Waste had a slightly higher, but B a slightly lower grade than f
those with 8.6 percent waste removed. However, differences of 1
1 percent in waste are probably not significant. a

The percentages of waste found in the cottons within each 5
composite grade are given fo-r each classer in Table 13. Although 5
there sometimes were wide differences in the ranges of waste, .
the averages for the 2 classers within each grade did not a
differ greatly. . “T

The classers differed more in staple of the uncleaned samples 
than in grade. Classer A gave a shorter average staple for ‘i.
each method of harvesting than classer B. Classer A found the ;
average length of the 132 uncleaned cottons was 26.8 in thirty- .;

Table 13. Waste falling within each grade for uncleaned cottons

Compoite grades for uncleaned samples

Grade ' Grade Grade Grade Grade 

4and 4% Sand 5% Sand 6% 7and 7% 8and8 '

Waste Waste ' Waste Waste Waste

Classer Range Av. Range Av. Range Av. Range Av. Range l
A . . . . . . . . . . . . . . . . . . . . . . . .. 4 0— 6 6 4 8 3.8—l3 7 7 0 5 7-17 6 9.2 5.6—23 2
B . . . . . . . .. 3 8— 6 6 4.7 4 l—l2 3 5 9 3.8—l6 0 7 3 5 7 17.6 9.8 l0.3—23 2

  
  
 
  
 
   
  
   
  
 
  
 

l EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 25

nds of an inch, but classer B found the average length was
f a difference of nearly 3/32 inch.

ey differed more than 2/32 inch for the uncleaned hand-
ed and snapped cottons and nearly 3/32 inch for the ma-
fe-picked cottons. The greatest difference between classers
for the stripped cottons, nearly 4/32 inch.

sser A gave shorter staples than B for 118 samples, the
_e for 12 and longer for 2 of the 132 uncleaned cottons.
ser A found that 39 cottons were 3/4 inch, but B only 3
ble 14). Classer A thought only 2 cottons were as long as 1
 but classer B thought 12 were 1 inch and that 6 were
er. Classer A found that 10 cottons were from 15/16 to
ch, but B found 73 were within this range.

‘e l4. Frequency distribution of staple for uncleaned and cleaned
 cotton as estimated by 2 classers

Uncleaned Cleaned
Classer
Staple, 32nd inch
A B A B
. . . . . . . . . . . . . . . . . . . . . _ . . . . . . . . . . . . . .. 39 3 l4 

IIIIIIIIIIIIIIIIIIIIIIIIK111111111"'19 ' ' ' ' ' ' "i6 ' ‘ ' ' ' ' "ii ' ' ' ' ' ' ' "12""

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII""49 ' ' ' ' ' ' "i7 ' ' ' ' ' ' "i3 ' ' ' ' ' ' "iv"

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15 23 22 21

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 6 42 2 56

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2 l9  l9

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2 12  9

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3  l

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3 

    
  
 
 
   
   
   
  
 
   
   
  

',he classers agreed on the staple of only 12 cottons (Table
. They differed 2/32 inch for 35' cottons, 4/32 for 15 and
2 for 16 cottons. They differed from 2/32 to 6/32 inch in-
give for 103 of the uncleaned samples.

Qe and Staple of Cleaned Cotton

a e 2 classers differed widely in grades for the cleaned sam-
. There was a difference of 2.2 grades between the average
the composite grades given by the 2 classers for the 132
ned samples (Tables 1 and 6). Classer A gave an average
4.5, or M-|-, and classer B an average grade of 2.3, or ap;
ximately GM+. The least difference between the classers
the different methods of harvesting was 2.1 grades for the
(hand-picked cottons and the greatest was 2.7 grades for
. 9 machine-picked samples.

or only 3 samples were the composite grades identical, in
trast with 55 for the uncleaned samples (Table 12). The
sers differed by 1.5 grades for 30 samples, by 2.5 grades
‘127, by 3.5 for 21 and by 4.5 grades for 1 sample. There was
ifference of 2 or more grades between classers for 82 of

26 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
   
   
 
 
 
   
  
  
   
   
     

the 132 cottons in contrast with only 2 samples of the uncleaned 
cotton.

With practically all the foreign matter removed, it seems 
that grade for leaf should be very high, perhaps grade number 1
2(SGM), if not grade number 1(MF). Classer B found the»
average grade for leaf to be 1.5(SGM+), but classer A only 3.3, 11
approximately SM+, a difference of nearly 2 grades (Tables 1 l“
and 6). The question arises as to whether classer A gave the 
cleaned samples too low a grade for leaf. Perhaps he was in- i
fiuenced by the recollection that none of these. samples in the un- 
cleaned state was above Middling, grade 5. 

The averages for the cleaned samples representing the 4 I
methods of harvesting, with wastes removed ranging from f
5.9 to 10.5 percent, differed a maximum of 0.4 grade for leaf »
for classer A and 0.3 grade for B (Tables 1 and 6). Therefore, s
each classer found little difierence in the average leaf grades f
of the cleaned samples regardless of previous waste content. 

The differences between classers for the average grades 1
for leaf range from 1.6 for snapped to 1.9 grades for machine- 5 ‘
picked. Individual leaf grades of 2, 3, 4 and 5 were included for % a
classer A, and grades 1, 2 and 3 for classer B. '2

In grades for leaf, the classers agreed for only 5 cleaned
samples (Table 12). They differed by 1 grade for 45, by 2E
grades for 65, by 3 grades for 15 and by 4 grades for 2 sam- 
ples. Of the 132 cottons, the classers differed by 2 or more leaf .
grades for 82 cleaned samples, but for only 3 of the uncleaned~i
samples. "

For classer A, the average grades for color ranged from 4.6;
for the 59 hand-picked cottons, from which 5.9 percent waste;
had been removed, to 5.5 for the 37 stripped samples from I
which 10.5 percent waste had been removed (Table 1). The:
average grades for color of the cleaned cottons were in the same,
relative po-sition with respect to former waste content, as were
the composite and leaf grades for classer A. Classer B found, :
however, that the samples which had an average waste of 8.63
percent removed were slightly higher in grade than those with7
only 5.9 percent removed. There was 0.5 grade difference be-g
tween the cottons which had 9.4 and those with 10.5 percent;
waste removed.

There were greater differences between the classers for;
color grades than for leaf grades. Classer A assigned an average
color grade of 5.0(M), in contrast with 2.7(GM+) assigned;
by classer B to the 132 cleaned samples. The classers differed;
3 grades for the machine-picked cottons. if;

The classers assigned identical color grades to only 4 sam.
ples (Table 12). They differed by 1 grade for 26 samples, by;

   
 
  
    
 
   
  
   
  
   
  
 
 
  
 
  
 
 
   
   
 
   
  
   
  
  
   
  

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 27

 aides for 51, by 3 grades for 30 and by 4 grades for 19
i468.

in determining the grade for color, the classer judged the
ibined effect of leaf and cotton in the uncleaned cotton,
* of necessity, only the cotton fiber in the cleaned state,
U the color of the fiber in the uncleaned state is not known.
fnly the color of the fiber had been considered in both the
eaned and cleaned states, it seems unlikely that the removal
 oreign matter would have changed the color of the fibers
selves as much as the classers found, from grade 6.5 to
l e 5 for class-er A and from grade 6 to grade 2.5 for classer
j; is possible that classer B gave slight attention to the color
 cleaned fiber and assumed that since the leaf grade Was
p , the grade for color was also high. This possibility is in
ement with the frequent observation that classers tend to
Are color unless a cotton is obviously stained or discolored,
7 color is one of the 3 factors included in grade. It, presum-
,, should receive as much attention as leaf and preparation.

 e classers differed by 2 or more grades for the composite
3e of 82 cleaned cottons, for leaf grade of 82 and for color
,~ of 100 cottons (Table 12). This is in contrast with 2
p ns for composite grade, 3 fo-r leaf and 5 for color for which
f classers disagreed by 2 or more grades for uncleaned cot-

3 e classers differed slightly more than 2/32 inch in the
rage staple of the cleaned samples. The least difference be-
y-- classers was for the snapped cottons, slightly less than
2 inclh, and the most was for the machine-picked, nearly
14 inc .

‘lasser A judged 14 cottons to be 24/32 inch and 21 to be
_ 2 (Table 14). Classer B found none as short as 24/32
~ only 8 to be 26/32 inch. Only 2 cottons were assigned staples
330/32 by A, but 56 Were given this length by B. Classer B
g d 29 to be from 31/32 to 33/32, but A found none longer
n 30/32 inch.

a he classers agreed on the staple of 14 cleaned cottons (Table
p. They differed 1/32 inch for 18, 2/32 for 57, 3,/32 for 27,
2 for 11, 5/32 for 3 and 6/32 for 2 cottons.

l e 15. Frequency distribution of differences between classers in staple

Difierence, 32nd inch Uncleaned cottons Cleaned cottons
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 14

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 18

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 57

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 27

, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l5 ll

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2

28 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

Classer A gave shorter staples than B for 236 of the 264 r
uncleaned and cleaned samples. They agreed 0n the staple of
26 of the 264 samples. ’

Changes in Grade and. Staple Due to Cleaning

The differences between classers in their estimate of the A
grades and staples of the cottons are reflected in the changes
in grade and staple due t0 cleaning. I

The changes in composite grade due to cleaning for indi- f
vidual cottons range from 0 to 5 grades for classer A and ‘
from 2 to 6 for classer B (Table 16). The average change 3
in grade for the 132 cottons was 2.1 for classer A and 4.0 1
for B, a difference of approximately 2 grades between classers
(Table 17).

Table 16. Changes in grade due to cleaning as judged by 2 classers

Number of cottons

 
    
  

Leaf grade I Color grade l Composite grade
Change
in Classer
grade
A B A B A B
0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 . . . . . . . . . . 3 . . . . . . . . . .

5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 . . . . . . . . . .

1 . . . . . . . . . . . . . . . . . . . 3 . . . . . . . . . . 55 l 18 . . . . . . . . . .

1 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 . . . . . . . . . . 12 . . . . . . . . ..

2 . . . . . . . . . . . . . . . . . . . 22 2 29 15 45 3

2 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 . . . . . . . . . . 16 9

3 . . . . . . . . . . . . . . . . . . . 51 10 16 51 20 l6

3.5 . . . . . . . . . . . . . . . . . 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 29

4 . . . . . . . . . . . . . . . . . . . 42 35 5 46 7 19

4.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

5 . . . . . . . . . . . . . . . . . . . 12 56 . . . . . . . . . . 19 1 l9

5.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

6 . . . . . . . . . . . . . . . . . . . 1 29 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Table 17. Frequency distribution of changes in staple due to cleaning and {
differences between classers 

Number of cottons
Classer
Change, 32nd inch Diflerence between
A B classers
0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 47 31

1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 61 45

2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 22 22

3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 l 14

4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 1 l3

5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 . . . . . . . . . . . . . . 6

6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Classer A found that the composite grades for 37 samples
were improved by cleaning from 0 to 1.5 grades, but classer B 
found that no cotton was improved so little, but that all sam- 1*
ples were improved by 2 or more grades (Table 16). For y.
classer A, 111 cottons were improved by cleaning from 1 to j
3 grades, while for classer B, 113 samples were improved 3 .
to 5 grades. Classer A found that 3 samples were not im- ’

   
 
 
  
 
  
 
  
 
  
  
  
 
 
 
  
 
  
   
  
 
  
 
 
  
 
 
  
  
 
  
 
 
 
  
  
 
  
  
  
 
  
  
  

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 29

ioved by cleaning and 4 samples by only 0.5 grade. Classer
J-fOUIId 45 samples to have been improved 2 grades, in con-
‘ t with only 3 samples according to classer B, the minimum
provement reported by B. The maximum improvement found

classer A was 5 grades, but for only 1 sample. Classer B
 nd 26 samples to be improved 5 and 6 grades.

Since the composite grade in this study is composed of the
Vbmed effect of leaf and color, the separate grades for leaf
m for color should show wherein the classers found their
eatest disagreement.

Since all of the cleaned samples were almost completely de-
fid of foreign matter, it seems that each cotton would have
Q- eived a high grade for leaf and that each would have had a
iuch higher grade than when uncleaned. Also, since the classers
» ered little for the average leaf grades of uncleaned cottons
.3 grade), it might be expected that the classers would have
 eed rather closely for leaf grade of the cleaned samples
f: thus would have found approximately the same improve-
gnt in grade due to the removal of foreign matter. The
sers did not agree closely, however, upon the leaf grades
A the cleaned samples, therefore, did not agree on the extent
y improvement due to the removal of foreign matter.

I The leaf grades given to the individual cleaned samples
nged from grade 2(SGM) to grade 6(_SLM) for classer A
"an from grade 1(MF) to 3(GM) for classer B. The improve-
ent in leaf grade due to cleaning ranged from 1 to 6 grades
r classer A and from 2 to 6 grades for classer B (Table 16),
'th averages of 3.3 and 4.8 grades, respectively (Tables 2

In 7).

 The greatest improvement in leaf, according to classer A,
 in snapped cotton, which was improved 3.7 grades. Classer
é found the greatest improvement in machine-picked samples,
' hich were improved an average of 5.4 grades (Tables 2 and 7).

lYThe classers disagreed by 1.5 grades for the average change
1 leaf grade of the 132 cottons. They disagreed 1.3 grades
or hand-picked, 1.7 for machine-picked, 1.3 for snapped and
.6 for machine-stripped.

 For 18 samples, the 2 classers agreed on the extent of the
1 provement in leaf grade following cleaning. For 94 samples
g ey differed by 1 and 2 grades, and for 19 by 3 and 4 grades.

1' It appears that the classers must have considered some fac-
frs other than foreign matter in assigning leaf grades to the
“leaned samples, or all cottons would have been more nearly
he same grade, since they were almost completely devoid of
oreign matter. If grade for leaf is based only upon foreign
atter content, and the samples were all free of foreign mat-

3O BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

ter, then they might be expected to be classed as MF, grade 1, l
or SGM, grade 2, for leaf. It seems possible that classer A may "j
have been influenced by the color of the cotton and gave the a
samples a low grade for leaf as Well as for color. It seems pos- i
sible that classer B, upon finding the samples free of foreign a
matter, gave them a high leaf grade and assumed that the
color was also good and, therefore, gave a high grade for color. i.

The 2 classers disagreed more widely on the change in grade ‘
of color produced by the removal of foreign matter than they 4‘

did on leaf grade.

For the 132 cottons, classer A found the grade for color to -l
have been improved an average of 1.3 grades While classer B l
found the change to be 3.5 grades, a difference between classers 
of 2 grades. Classer A found that the cottons for 2 of the 4 
methods of harvesting were improved 1.4 grades, 1 by 1.2 f
grades, and the fourth method, machine-picked, by 1.6 grades. .
Classer B found the improvement to range from 3.3 for hand- =

picked to 4.4 grades for machine-picked cottons.

Classer A found that there was no change in the color of 
24 samples, but an improvement of 1 grade for 55, 2 grades i-
fo-r 29, 3 grades for 16 and 4 grades for 5 samples (Table 16). *
Classer B found that only 1 sample was improved so little as j,
1 grade. Other changes for B were improvements of 2 grades 

for 15, 3 for 51, 4 for 46 and 5 grades for 19 samples.

From these differences between classers, it appears that the =
same factors were not considered in assigning grades for color. 1
The question ‘comes to mind as to how much attention each L
classer gave to the. area and color of the foreign matter, how .
much to the color of the fiber, and how much to the combined ¢

effect of the color of foreign matter and fiber.

Nickerson“ found by measuring the color of uncleaned cotton 1
(both fiber and foreign matter) and the color of the cleaned 
cotton, that the grade was raised by 1 to 2.5 grades. The dif- 
ferences found by classer A, an average for the 132 cottons of _

1.4 grades, fall within this range.

It seems possible that classer B, associating clean cotton with 
a good grade for color, gave too high a grade for the color of ;
the cleaned cottons, which resulted in an apparent improvement 

of 3.5 grades.

In only 1 of the 4 lots of cotton, machine-stripped, did classer 
A find an average difference in staple, due to cleaning, of more 5
than 1/32 inch. Average differences for all 4 methods of har- 
vesting were insignificant for classer B and for 3 methods for f;

classer A. a

3Effect of Cleaning on Grade and Color of Cotton, by Dorothy Nickerson. 

USDA, PMA, Cotton Branch, April 1947.

   
   
 
   
  
  
    
   
   
 
 
  
  
  
  
   
   
  
 
  
  
  
   
 
  
    
    

 EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON s1

'r individual cottons, however, the classers frequently found
'derable difference between the staple of the uncleaned
the cleaned samples. For example, classer A found that 28
les changed from 3/32 to 5/32 inch, inclusive (Table 17).
e_r 113 found that 24 samples changed from 2/32 to 4/32
. 1nc usive.

e removal of waste had no effect upon the length of 46
les for classer A and none on 47 for classer B, but only
’f these were the same samples. Classer A found that 54
les gained in length and classer B found that 41 gained,
g which were included by both classers. Losses were re-
2'0 fo-r 32 samples by A and for 44 by classer B, 16 of
h were the same cottons. For only 31 of the 132 cottons
the classers agree as to the degree of change in length
to the removal of waste. The classers differed in their esti-
on of change in length by 1/32 inch for 45 cottons, by 2/32
1i 2, by 3/32 for 14, by 4/32 for 13, by 5/32 for 6 and by
 inch for 1 cotton (Table 17).

fin differences between classers for the changes in grade and
le due to cleaning are reflected in the prices per bale which
_d have been received using each classification. It should
ept in mind that the grades and prices used, the only ones
able, are those designed for ordinary ginned cotton with
1.4 foreign matter.

we average price per bale of the 132 uncleaned cottons on
basis of the classing of A was $128.90. The average price
Pbale was $144.48, using the classifications of classer B.
f» e was a difference of $15.58 per bale due to the differences
een the classers’ estimates of the grades and staples of
uncleaned samples. This difference was due chiefly to the
rence in their estimates of staple.

sser B found that the cottons harvested by each method
 Worth more per bale, both uncleaned and cleaned, than
classer A. The average differences between the classers
gthG 132 uncleaned cottons, cottons with which both classers
'- familiar, were $15.28 per bale. The differences between
‘ers were $15.03 per bale for the hand-picked, $12.45 for
Fmachine-picked, $11.05 for the snapped and $20.53 per bale
1 the stripped.

he differences between the 2 classers in their classifications
 e 132 cleaned cottons, with which they were not familiar,
lted in a difference in the average price of $11.47 per bale,
ch is less than for the uncleaned samples. The differences
§_een classers in prices of cleaned cottons were $10.93 per
 for hand-picked, $12.69 for machine-picked, $11.27 for
,ped and $12.16 per bale for stripped cottons. The classers

32 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

differed more in the average price per bale for the uncleancd 
samples, with which they Were familiar, than for the cleaned 

samples, with which they were not familiar.

Classer A found that 17 and classer B that 20 cottons were 
sufficiently reduced in weight by cleaning to more than offset
the gain in price due to the higher grade. Only 4 of the cottons 

which thus lost were included by both classers.

The price changes per bale Were from a loss of $20.98 to a 
gain of $48.63, a range of $69.61 for classer A, and from a 
loss of $15.64 to a gain of $34.83, a range of $50.47 for classer 1
B. Classer A found that 96 cottons gained and 36 lost in price, ;

and B found that 82 gained and 50 lost by cleaning.

The average increase in price of the 132 cottons due to clean- 
ing was $10.21 per bale, according to classer A, and $6.09, ac- ii
cording to classer B (Tables 2 and 7). That the average changes 
in price are so nearly the same for the 2 classers, despite the ‘
relatively wide differences in grades, is due in part to the l
fact that there were no premiums given fo-r grades above GM, i“
grade 3, on the date of the prices used in the calculations. For 
these reasons those 96 cleaned cottons to which classer B as- 7

‘ signed grades SGM, grade 2, and MF, grade 1, received no high- 

er price than those which were classed as GM, grade 3. With T
our present system of grading and at these prices, there was E
no advantage to the grower in attempting to produce cotton 

which graded above GM, grade 3.

Despite the small difference between averages for the 2_'c_lass- l
ers there are wide differences between classers for individual 

Table 18. The gain or loss in price per bale with the we-ight 0f waste 

removed by cleaning

No. of cottons and av. waste No. 0f cottons and av. waste‘.
in lbs. if price was a loss in lbs. if price was a gain
Classer Classer
Change in price ~
per bale A B A B
No. Lbs. No. Lbs. No. Lbs. No. Lbs.
$ $

0.00— 2.99 . . . . . . . . .. 9 38.8 12 36 6 ll 32.7 9 39.3

3.00- 5.99 . . . . . . . . .. ll 34.6 13 27.0 14 42.3 8 52.5

6.00— 8.99 . . . . . . . . .. 8 30.0 15 31.4 8 50.4 7 47.6

9.00— 11.99 . . . . . . . . . . 1 26.0 5 25.8 6 45.4 l0 45.8

l2.00— 14.99 . . . . . . . . . . 3 27.8 3 38.5 7 45.4 17 45.8

15.00-— 17.99 . . . . . . . . . . . . . . . . . . . . . . . . 2 27.5 8 50.7 13 45.2

18.00— 20.99 . . . . . . . . .. 4 27.1 . . . . . . . . . . . . .. 7 40.4 5 45.4

2l.00— 23.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 42.3 5 50.8

24.00— 26.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 42.1 1 59.0

27.00— 29.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 48.6 1 45.0

30.00— 32.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3 44.3

33.00—- 35.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. l 52.0 3 46.2

36.00— 38.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7 36.1 . . . . . . . . . . . . .. 

39.00— 41.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .; . . . . . . . . . . .  "

42.00—- 44.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2 51.8 . . . . . . . . . . . . .. ‘

45.00— 47.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 1 50.0 . . . . . . . . . . . . ..

48.00— 50.99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 1 45.0 . . . . . . . . . . . . ..

  
 
  
 
 
  
 
 
  
  
 
 
 
 
 
  
 
 
  
   
 
  
   
  
 
  
 
  
  
 
 
  
  
 
 
  
 
 
    
    
 
 
  

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 33

 ples. The number of cottons for each classer which fall
fithin the different ranges 0f price changes is given in Table
‘ 8. For 86 samples classer A found the changes to be less than
 while B found 114 to be within this range. For A, 35,
‘nd for B, 18, fell between $18 and $36. Classer A found that
_,1 cottons changed more than $36 per bale with a maximum
ange of $48.63, while B found no cotton changed more than

. 4.83.

“gj Classer A found that the greatest loss due to cleaning was
20.98, falling within the range of $18-$20.99, as given in Table
f8. Classer B found the same cotton gained $2.43, a difference
' $23.41 between classers. B found that the greatest loss was
 15.64, while for the same cotton A found a loss of only $6.09,
’ difference of $9.55 between classers. The greatest gain was
8.42 for classer A, and for B this cotton gained $32, a dif-
‘A rence of $6.42. For B the greatest gain was $34.83 for which
found a gain of only $22.90, a difference between the classers
3» $11.93. The greatest difference between the classers for any
tton was $42.46 for a cotton which A thought gained $4.62

y cleaning and B $47.08.

g One cotton lost $1.15 by cleaning, according to one classer,
I d gained $36.41, according to the other classer, a difference
g $37.56. In another case, the grower would have gained
23.67 or lost $1.36, according to which classer’s classification
as used. In 55 cases the differences between the classers in
nges in price per bale were between $10 and $42.46, 25 of

‘hichwere above $20. In only one case did the classers agree

',| the change in price, a gain of $5.43 per bale, but in 39

=ses they differed less than $5 per bale.

I With the 2 classers differing so much in their estimates of
i. effect of cleaning upon the grade and staple of these 132
ttons, it is difficult to draw conclusions from these data as
I the probability of the grower profiting by the complete re-
oval of foreign matter, if equipment were available. It seems
i- obable, however, that it would not be profitable to remove
reign matter from cotton which grades SM or better and
‘hich contains no more than 5 or 6 percent foreign matter.
or cottons of lower grade and with a higher percentage of
‘reign matter, the removal of waste would probably be profit-

v l6.
I Other Studies

In a preliminary study of 60 cottons classed by a board of
miners both before and after cleaning, it was found that
y. removal of foreign matter raised the grade approximately
j grades. Following the report of this study, Nickerson reported
study of the same problem.

It. should be pointed out before comparing results of the 2

34 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

studies that the manner of classing used for the Nickerson study
differed in one very important point from that employed in the
Texas studies. For the cottons included in the Nickerson study
the author stated that “paired samples of cleaned and uncleaned
lint were prepared in boxes like those used for the cotton stand-
ards for grade.” This pairing of samples made it possible for the ‘
chairman of the Washington Appeal Board of Review Exam-
iners, Who classed the cottons, to class the uncleaned along
with the cleaned sample. The author points out that “this
severe contrast sometimes made it necessary to lower the
grade originally given to the ginned lint cotton.” Theoretically,
in classing each cotton the grade is determined by comparing
it with the standards and not with some other cotton.

The classers who classed the cottons for this Texas study
had no such opportunity to pair samples or change the grade
of one of the pair to conform with the grade of the other.
The cleaned and uncleaned samples were coded so that the
classers would not be tempted to check one against the other
but would presumably check each sample against the stand-
ards. These classers were given no information as to the source
of the cotton or the method of harvesting used, but all of the
cottons were grown in Texas and were those with which they
were familiar. They were told, however, that the cleaned sam-
ples were the same as the uncleaned, but with the foreign
matter removed with a Shirley analyzer. It was thus thought
an unbiased report would be obtained, free of any preconceived
notions or personal prejudices.

A small number of College Station and Lubbock samples
were included in the Nickerson report, making it possible to
compare the results of a portion of the 2 studies. Nickerson
reported that the grade of College Station hand-picked sam-
ples when cleaned were improved an average of 1.5 grades, 2
grades for snapped, slightly over 1 grade for machine-picked
and 2 grades for stripped cottons. In the present study, ac-
cording to classers A and B, respectively, hand-picked samples
from College Station were improved 2.4 and 3.8 grades, snapped
2.4 and 4.4, machine-picked 2.0 and 3.8 and stripped 2.4 and
3.8 grades. Differences between the 2 studies for classers A
and B are respectively, approximately 1 grade and 2.5 grades
for the hand-picked samples, 0.5 and 2.5 grades for snapped,
0.5 and 2.5 for machine-picked, and 0.5 and 2 grades for stripped.
Only snapped and stripped cottons from Lubbock were com-
mon in both studies. Classer A agreed closely with the average
improvement reported by Nickerson, approximately 2 grades.
Classer B found improvements of 4.4 and 3.8 for snapped
and stripped samples, respectively, a difference of approxi-
mately 2 grades from the Nickerson study".

From color measurements of the cottons converted to grade, . fi
Nickerson found the improvements in grade for color at Col- 


   
   
 
 
  
 
 
 
  
  
  
 
  
   
 
  
  
   
 
   
   
  
 
  
  
 
  
 
  
  
 
 
 
  
  
 
 
 
  

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 35

 Station were for hand-picked 1.5, snapped 2, machine-
;ed 2 and stripped 2.5 grades. For the present study, classers
feand B found that College Station hand-picked samples were
troved in color approximately 2 and 3.5 grades, respectively,
 hine-picked 1.5 and 3.5 grades, snapped 1.5 and 4 grades
,~ stripped 1.5 and 3 grades. It appears that classer A dif-
;_d from the Nickerson study by only approximately 0.5
de and classer B from 0.5 to 2 grades for color in the
' int of improvement by cleaning.

3 or the Lubbock snapped and stripped cottons, classer A’s
,3 'mate of improvement approached that reported in the Nick-
on study, but classer B differed by 2.5 grades.

%The cottons reported in the Nickerson study were taken from
' annual variety studies. They contained less foreign matter
y: those reported in this study. The cottons used for the
ckerson study were probably most, if not all, ginned on
.;. ll laboratory gins. Many of those in the present study were
fned on large commercial gins which remove a higher per-
tage of burs and fine trash than the single-breasted labora-
' gin. The differences in ginning and consequent foreign
tter may account in part for differences between the 2
 dies. ‘
Summary and Conclusions

Samples of 132 cottons representing different varieties, meth-
i.» of harvesting and locations were classed by 2 classers,
th before and after all foreign matter was removed by a
irley analyzer, to determine, if possible, the effect of foreign
_tter and its removal upon the grade, staple and price.

ftThe classers were also sent duplicate samples. One checked
gluplicates out of 5 cottons for grade and 1 of the 5 for staple.
a e other classer agreed on the duplicates for 3 out of 10
ttons for grade andon 2 for staple.

"Classer A found that cleaning improved the grade for leaf
'33 grades, the color 1.3 grades and the composite grade 2.1
l ades. Classer A found that 96 cottons gained and 36 cottons
j,» in price due to cleaning. The price was increased by clean-
g- an average of $10.21 for the 132 cottons.

f Classer B found that cleaning improved the grade for leaf
,8 grades, color 3.5 and the composite grade 4.0 grades. Clean-
: g increased the price per bale of 82 cottons and decreased the
rice of 50 cottons, according to classer B. The price was in-
l» eased by cleaning an average of $6.09 per bale.

_ Measurements of the length of the 264 uncleaned and cleaned
‘mples with the Fibrograph agreed closely with those of classer
_. The Fibrograph showed no significant effect upon the aver-

    
    
     
  
  
 
    
  
   
   
 
 
    
     
 
    
   
   

36 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION

age staple by cleaning. Neither classer found that cleaning sig
nificantly changed the average staple, although short-fibere
cottons appeared to be slightly longer and long-fibered cottons
slightly shorter after cleaning. F

Methods of harvesting caused less than 0.5 grade differen I?
in the improvement due to cleaning, according to classer A
Classer B found the greatest difference in improvement be-g
tween hand-picked and machine-picked cottons, 1.1 grades. I:
found that machine-picked cotton showed the greatest improve-qt
ment, although it contained slightly less waste than snapped
or stripped cotton. T

The effect of location on the improvement due to cleaning
varied with the classer. For example, one classer found stripped}
cottons from College Station were more improved by clean-
ing than those from Lubbock, while the other classer found,
the reverse was true. T

The 2 classers differed less than 0.5 grade on the average,
grade of the 132 uncleaned cottons. They agreed on 55 samples‘
but differed from 1 to 2.5 grades for 34 samples. ’

The 2 classers differed slightly more than 2 grades on the;
average grade of the 132 cleaned cottons. They differed 2 orf_
more grades for the leaf grade of 82, the color grade of 100,
and the composite grade of 82 cleaned cottons. v

The classers differed by nearly 3/32 inch in the averages
staple of the uncleaned and slightly over 2/32 inch for the
cleaned samples. Neither classer nor the Fibrograph found?
that cleaning changed the staple significantly. '

The classers differed only $4.12 per bale for the average
improvement in price due to cleaning but differed widely fori
individual cottons. They agreed upon the direction of the change
for 95 of the 132 cottons. They differed from $10 to $42 in.
the extent of change for 55 cottons. Classer A found that 96*;
cottons gained and 36 l.ost by cleaning, while B found that 82,
gained and 50 lost. >

The same cottons which, according to classer B, lost $1.15,?
$2.30, $0.92, $5.53 and $8.93 per bale by cleaning gained?
$36.41, $29.20, $22.86, $23.48 and $25.92, respectively, accord?
ing to classer A. On the other hand, cottons which lost $18.90.,
$19.11, $1.36, $2.50 and $6.13 per bale, according to the classifi-i,
cation of A, would gain $1.34, $0.10, $23.67, $14.50 and $15.85
per bale, respectively, by .B’s classification. These differences;
between classers’ judgments would cause differences in pric
per bale for these same cottons of $17 to $38. a

The extent to which the classers agreed apparently bore no
relationship to the amount of foreign matter in the uncleaned

EFFECT OF CLEANING ON GRADE, STAPLE AND PRICE OF COTTON 37
O

lint as there were differences greater than $10 for cottons with
wastes ranging from approximately 4 to 23 percent.

It is doubtful if the increase in price of relatively clean cotton
through the removal of foreign matter would be great enough
to compensate for the loss in weight. On the other hand, cotton
with much foreign matter, although it loses a good deal of
weight by cleaning, is raised enough in grade to materially
increase the price. It appears probable, therefore, that cottons
with a small amount of foreign matter will not bring more when
cleaned, but that cottons with considerable foreign matter can
be cleaned profitably.

The frequent wide differences between the 2 classers in their
assignment of grade and staple are probably caused by a
number of factors. From examination of the uncleaned and

g cleaned samples, it seems that classer B paid too little atten-
it tion to the color of the cleaned samples. On the other hand,
 classer A surely graded the leaf of the cleaned samples too
l low, for they had practically no visible foreign matter, there-
fore, might be expected to have an average leaf grade of SGM
or better instead of SlVI-k. Classer A apparently staples All
 cottons shorter than does B. In only 3 of 264 cases (132 un-
; cleaned and 132 cleaned samples) did B assign a shorter staple
‘j than A, while in only 26 cases did they agree on the staple.

   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
    
  
  
  
    
 
  

1 Classers apparently are more concerned with the area occu-
T. pied by foreign matter than with the kind and weight of the
 foreign matter. This study supports the frequent observation
p that classers tend to ignore color unless a cotton is very ob-
. viously tinged or stained.

_ The results of this study emphasize anew the dependence of
1 the grower on the ability and fairness of the classer. Despite
 the provision of standards, well-trained, experienced, conscien-
tious, licensed classers are found to differ not only from each
other, but to differ from sample to sample of the same cotton.

, The prices paid to farmers are based on government stand-
jards for grade. The standards used by individual firms and
‘classers frequently are modifications of government standards.
The use of difierent standards, plus inevitable human errors,
may result in wide differences between the grades assigned
i_ by classers and may mean a material loss to the farmer. Gov-
ernment grade is the chief determinant of price, therefore, it
lis extremely important that each sample is graded according
jto actual government standards.

" It is recommended that classers use every opportunity to
~ check their grades and staples with each other and with actual
hysical measurements of the physical properties. It is en-
ouraging to learn that an increasing number of classers are
ow following this practice, which should result in a more just
deal for all concerned with cotton.

38 BULLETIN 723, TEXAS AGRICULTURAL EXPERIMENT STATION
I

Acknowledgments

Grateful acknowledgment is made to the classers who classed 
the cottons and to the Cotton Branch, Production and Market- 
ing Administration, U. S. Department of Agriculture, which "j
permitted use of the Shirley analyzer in their cotton testing f;

laboratory.