THE EFFECT OF VAEEANGNS OF FILE STRUCTURE
@N THE CQLGR RED-WERE

Thesis {re-v {‘{m Duran a»; M. A.
MECHEGAN STATE UREVERSITY

Anita. Marie Fratime
1965

C - u‘ -“--'.~...-

 

     
     

LIBRARY }
MiChigan State ,5
University

MlWI!mzllljflllfllflflflljflwWI ‘

lhula

 

 

ABSTRACT

THE EFFECT OF VARIATIONS OF PILE STRUCTURE
UPON THE COLOR RED-PURPLE

by Anita Marie Fratianne

This research was undertaken to investigate, in
detail, the effect of variation of pile structure upon the
color, red-purple. Effort was made to determine if a varia-
tion of pile height, yarn size, and pile density, and/or an
interrelationship of each, affect a pattern of color change
on red-purple, and if the behavior of the color red-purple,
when affected by texture, is related to its spectral composi-
tion. To answer these questions, samples were woven that
consisted of three sets of pile structures. One set of sam-
ples was woven with red—purple yarn, the color under investi—
gation; the two sets of control samples were woven, one of
red and one of purple yarn. The pile structure of each sam-
ple varied in size of yarn ply, pile height, and pile den-
sity. The combination of variables resulted in a total of
24 samples.

A subjective color analysis was made by three judges
by matching the color of each sample with a Munsell color
chip. As the pile height and pile density of the samples

decreased, the hues, generally, appeared to be warmer, and

Anita Marie Fratianne

the value and saturation levels of the colors decreased. As
yarn ply decreased, however, the value and saturation levels
of the colors increased.

The objective color analysis was done by the General
Electric Recording SpectrOphotometer. The data were then
substituted in the Nickerson and Stultz color difference
formula. Comparisons of the reflectance curves were also
made. The Nickerson and Stultz formula showed that the
color differences that resulted from a decrease in yarn ply,
pile height, and pile density were negligible. Comparison
of the reflectance curves showed reduction in size of yarn
ply, pile height, and pile density resulted in losses in the
percentage of reflectance in the red region of the spectrum.
In most cases, the reflectance curves of the red-purple sam-
ples more closely matched the curves of the red than the
purple samples. As pile structure varied, the color of the
red—purple samples reflected nearly the same kind and qual-
ity of light as did the red samples.

As pile structures were varied, no regular pattern
of color change could be found for the color red-purple.
However, the basic hypothesis of this study was proven: the
structural variations of the pile samples produced greater
patterns of irregularity in red-purple than in either red or
purple. The data further indicated that perhaps the influ-
ence of the red in the color red-purple was responsible for

the difficulty with which red-purple was discerned.

THE EFFECT OF VARIATIONS OF PILE STRUCTURE

ON THE COLOR RED-PURPLE

BY

Anita Marie Fratianne

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Textiles, Clothing and Related Arts

1966

ACKNOWLEDGMENTS

The writer wishes to acknowledge her sincere appre-
ciation to Lorraine Gross for her kind assistance and encour-
agement in directing this research.

Gratitude is extended to Dr. Mary Gephart, Dr. Elinor
Nugent, and Mary L. Shipley for the helpful suggestions and
criticisms that they have offered throughout this study.
Further, the author wishes to acknowledge Jeanne Myskins,
Lorraine Gross and Robert Bullard, the three judges in the
study who gave so freely of their time and talent.

Special gratitude is given to the author's family
and friends who have offered their patient, understanding
encouragement throughout the extent of this research.
Finally, deep appreciation is expressed to the author's
future husband, Dr. Leverett J. Zompa of the Chemistry
Department, Michigan State University, for his continuing
faith and support. Without his willing assistance this

research could not have been completed.

ii

TABLE OF CONTENTS

Chapter Page
I. INTRODUCTION . . . . . . . . . . . . . . . . 1
Statement of the Problem . . . . . . . . 1
Focus of the Study . . . . . . . . . . . 2
Review of Literature . . . . . . . . . . 3
Summary . . . . . . . . . . . . . . . . 21
II. METHODOLOGY . . . . . . . . . . . . . . . . 23
Research Design . . . . . . . . . . . . 23
Analysis . . . . . . . . . . . . . . . . 26
III. SUBJECTIVE COLOR ANALYSIS . . . . . . . . . 33
The Effect of Variation of Yarn Ply
on Red, Red-Purple, and Purple . . . . 33
The Effect of Pile Density on Red,
Red-Purple, and Purple . . . . . . . . 37
The Effect of Pile Height on Red,
Red—Purple, and Purple . . . . . . . . 41
The Effect of Variations of Pile
Structure on Red-Purple . . . . . . . 45
Summary . . . . . . . . . . . . . . . . 45
IV. OBJECTIVE COLOR ANALYSIS . . . . . . . . . . 47
Color Difference Formula . . . . . . . . 47
Reflectance Curves . . . . . . . . . . . 52
Summary . . . . . . . . . . . . . . . . 55
V. COMPARISON OF THE SUBJECTIVE AND OBJECTIVE
METHODS OF COLOR ANALYSIS . . . . . . . . 56
VI. SUMMARY AND RECOMMENDATIONS . . . . . . . . 61
Summary . . . . . . . . . . . . . . . . 61
Recommendations . . . . . . . . . . . . 65
BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . 67
APPENDIX: Reflectance Curves . . . . . . . . . . . 70

iii

LIST OF TABLES

Research design for samples: combination
of variables . . . . . . . . . . . . . . . .

Subjective color analysis showing the effect
of variations of yarn ply on red, red-purple,
and purple . . . . . . . . . . . . . . . . .

Subjective color analysis showing the effect
of variationscfifpile density on red, red-
purple, and purple . . . . . . . . . . . . .

Subjective color analysis showing the effect
of variations of pile height on red, red-
purple, and purple . . . . . . . . . . . . .

Color difference between samples that differ
from one another in either size of yarn ply,
pile density or pile height. . . . . . . . .

Objective color analysis showing comparison
of reflectance curves of the effect of
variations of pile structure on red, red—
purple, and purple . . . . . . . . . . . . .

Comparison of the responses of the three

judges and the spectrophotometer translated
into Munsell nomenclature . . . . . . . . .

iv

Page

24

34

38

42

48

53

57

CHAPTER I
IN TRODUC TION
Statement of the Problem

Texture and color are tremendously important to the
designer. Knowledge of the effect of texture upon color
helps the designer predict the influence of textural change
upon the apparent attributes of color, hue, value and satu-
ration.l

The subject of color modification by texture has
been especially interesting to researchers in the Textiles,
Clothing, and Related Arts Department at Michigan State
University.

Lorraine Gross2 showed that texture has a modifying
effect upon color. She ascertained that textural change

caused perceivable and measurable color changes that could

be expressed in Munsell color units. She noted that a

 

lHue, value and saturation is the nomenclature of
the Munsell Color System that describes the three attributes
of color. Hue is the name of the color, value is the light-
ness or darkness of the color and saturation is the bright-
ness or chromaticity of the color.

2Lorraine Gross, "The Effect of Height and Yarn Size
of a Pile Texture on Color" (unpublished Master's thesis,
College of Home Economics, Michigan State University, 1961).

change from a tapestry weave to a pile weave as well as a
change in height and yarn size of a pile structure produced
color changes in the samples.

The textural changes caused the majority of the
colors in the Gross study to change in a consistent direc-
tion; however, the blue-green, yellow—green, and red-purple
samples exhibited a pattern of color change that deviated
from the rest. Gross offered the theory that hand dying had
been responsible for the irregular color changes. The yarn
of the red-purple sample was commercially dyed and the
researcher could offer no other explanation for the deviant
color change of the red—purple. Further research was recom-

mended to eXplain the unique behavior of the red—purple.l

Focus of the Study

 

The intent of this research is twofold: (l) to help
determine whether variation of pile height, yarn size, and
pile density, or an interrelationship of them, affects a
pattern of color change in red-purple, and (2) to help
determine whether the unique behavior of the color is re-
lated to its spectral composition. Because the color red-
purple is made of two wavelengths of color, red and purple,
the latter two colors were used in this study as control

colors.

 

Ibid.

The basic hypothesis of this research is that tex-
tural variations upon the color red-purple produce a greater
pattern of irregularity than is produced by a textural varia-
tion upon either red or purple. This research, while basic
in nature, may help establish a practical method to assist
designers in predicting the effect of a specific textural

change upon a color.

Review of Literature

 

Color Theory

The literature concerning color theory was studied
because of its relevance to the topic analyzed in this

thesis.

Color and Light

 

Color is the aspect of radiant energy of which
the human observer is aware through the visual
sensations which arise from the stimulation of
the retina of the eye.
Color exhibits such a multidimensional nature that
scholars with varied interests study color sensations within
the structures of their own disciplines. To the physicist,

color is a study of light; to the physiologist, color is a

study of vision; to the chemist, color is a study of dyes

 

1Ralph M. Evans, "Variables of Perceived Color,"
Journal of the Optical Society of America, Vol. LIV, No. 12
(December 1964), 1467-1474.

 

and pigments; and to the psychologist, color is a study of
sense perceptions.

Color is a function of the physical quality of light.
Visible light is a small portion of the electromagnetic
radiation band which includes all forms of radiant energy.
This band is divided into sections, each section being spec-
ified by difference in wavelength. The wavelength associated
with visible light falls in a region approximately midway in
the energy band.

Light, as does all electromagnetic radiation, travels
in transverse waves. These waves can vary in frequency of
oscillation, thereby giving characteristic colors.1 The
frequency may be eXpressed directly in terms of cycles per
second. Another method of eXpressing color is in terms of
the wavelength2 which is a measure of the distance between
wave crests.3 Since this distance is extremely small, the
measurement is usually given in units of millimicrons (sym-
bol me), 0.0000001 centimeters, or in Angstrums (symbol 8),

0.00000001 centimeters.4

 

lMaitland Graves, Color Fundamentals (New York:
McGraw-Hill Book Co. Inc., 1952), p. 5.

2Deane B. Judd and Gunter Wyszecki, Color in Busi-
ness, Science, and Industry (2d ed.; New York: John Wiley
and Sons, Inc., 1963), p. 28.

3W. D. Wright, The Measurement of Colour (London:
Helger and Watts, 1958), p. 1.

4Martin Koblo, World of Color, trans. Ian. F. Finlay
(New York: McGraw-Hill Co., 1956), p. 9.

The color spectrum ranges from blue light whose wave-
length is approximately 400 mfiL(4000 R) to red light whose
wavelength is about 700 mil (7000 R). All color perceived
by the human eye lies between 400 mil to 700 mil.

The divisions of the Spectrum occur at the following
wavelengths.l

92.22:. 2311

Violet . . . . . . 400-450
Blue . . . . . . . 450-500
Green . . . . . . 500-570
Yellow . . . . . . 570-590
Orange . . . . . . 590-610

Red . . . . . . . 610-700

Color and Texture

 

Color results from the selective absorption and
reflection of light at different wavelengths by the surface
structure of a material. When light encounters a medium,
some wavelengths of light are absorbed and others are re-
flected. Generally, it is the reflected wavelengths of
light which we perceive as the color of the material.

The characteristic of a material such as texture,
luster, or transparency have the capacity to reflect, re—

fract, and absorb light, thereby determining the perceived

 

1A. C. Hardy, Handbook of Colorimetry (Cambridge:
Technology Press, 1936), p. 2.

 

color to the degree that the surface structure modifies the
light which falls upon it.

When light falls upon the surface of a material and
encounters a boundary between media,1 for example, a ray of
light between air and a cotton fiber or between air and a
pigment particle, then the reflection of the light occurs;
next, refraction occurs when the ray of light bends as it
encounters the medium; and finally, reflection reoccurs when
the light passes from the medium back to the air. Four per
cent2 of the initial beam of light is reflected upon each
encounter of light and the medium. When light meets a gran-
ular or fibrous structure, the beam of light meets with new
interference every few millionths of an inch. These repeated
encounters result in a thorough diffusion of the light.

It should be emphasized that Specular reflections
occur at the surface of a material, and that diffuse reflec-
tions occur beneath the surface in materials that have a
granular or pigmented material structure. After specular
reflections occur, the remaining light penetrates the sur-
face of the granular or pigmented structure. It is absorbed

while scattering multiple reflections beneath the surface,

 

1Richard 8. Hunter, "Those Other Aspects of Appear-
ance," Color Engineering, Vol. 1, No. 2 (June 1963), 8-14.

2

 

Ibid., p. 8.

and is finally readmitted into the air in a nearly diffuse

state.1

Thus, the material structure of an object modifies
light and determines color by reflecting, refracting, and
absorbing light. Particle size, and distribution and con-
tinuity of pigment or fiber are important structural factors
of materials that modify light. The effects of such physi-
cal phenomena as material structure upon light are eXplained
by Luckiesh:

The character of the surface of a pigment and
the density of coloring matter influences the
appearance of color. If a red aniline dye
solution be deposited upon a fabric by means

of an air brush, the pigment under some con-
ditions will be deposited in the form of

fluffy powder whose appearance is deep velvety
red. The purity of the red is largely due to
the multiple reflections, for the light is able
to penetrate an appreciable distance into the
medium, and at each reflection it becomes purer,
that is, more nearly monochromatic. If the
solution were applied by means of an ordinary
brush, the deposit would not have been so porus
and the appearance of the color would not have
been such a deep red.2

Color and Vision

 

The appearance of color is a function of the physi-
cal process involving the retina of the eye and the brain,
as well as the observer's psychological interpretation of

the physiological response. The observer's color sensation

 

1Judd and Wyszecki, p. 97.

2Mathew Luckiesh, Light, Vision and Seeing(Newaork:
D. Van Mostrand Co. Inc., 1944), po 36.

is, therefore, dependent upon the quality of light, the
proper functioning of the eye and the brain, and the accu~
rate interpretation as a result of the observer's psycholog-
ical process.

The sensation of color is determined to a great
extent by eye function. Evans clearly recognized that "vari-
ations in the human eye from one observer to another are
simply part of the subject of color.”1 In the properly func-
tioning eye, the conversion of light energy into nervous
energy takes place in the retina of the eye. The actual
receptors are the rods and the cones, containing a photo-
chemical substance which decomposes because of the action of
the light and reforms in the dark. Wright states that rods
are responsible for the detection of weak light, and the
cones, which function under higher illumination, are respon-
sible for the perception of detail and color.2 An observer
with normal trichromatic color vision is able to perceive
basically red, green, and blue light. ,Such observers, as
Judd explains, may differ considerably in their ability to
make fine color distinctions but this is a difference in
degree not in kind of color discrimination.3 However, as a

result of talent and training, the eXperienced observer is

 

1Ralph M. Evans, Introduction to Color (New York:
Wiley and Sons, Inc., 1954), p. 117.

2Wright, pp. 28-29.

3Judd and Wyszecki, p. 71.

able to discriminate and recognize colors that often remain
hidden to the less experienced observer.l
When nervous impulses from the retina reach the

brain what color is seen also depends upon the consciousness
of the observer; and, as a result, a whole new series of
phenomena must be included in the subject of color vision.
Depending upon the color of the surrounding areas, an
observer with normal color vision may view a wavelength of
light, for example, blue light, as a different color than it
actually is. Evans says that the effect of surrounding
areas are just as much prOperties of the eye as any of its
other characteristics, but they are usually considered part
of psychology.2

The sense of vision is curiously involved in a

number of illusions and phenomena, due largely

to the fact that the eye does not merely record

the objects before it but actually creates

effects in and of itself. Color perception is

not determined solely by the amounts and quality

of the light rays that reach the eye, but in-

volves the total situation in which the color is

seen.
The psychophysical adaptations of the eye to one color, and
the resulting effect of the ability of the eye to perceive

the color's complementary hue, should be recognized as

important influences upon color perception. Such color

 

lKoblo, p. 16.

2Evans, Introduction to Color, p. 117.

 

3Frederick M. Crewdson, Color in Decoration and
Design (Illinois: Frederick J. Drake and Co., 1953), p. 48.

10

phenomena as successive and simultaneous contrast of colors
are also important influences upon color perception.
Fortunately, however, Wright has assured color
researchers that while these forestated influences upon
color must be considered, their effect upon accuracy of
color match remains nominal:
Experiments show that colour match remains

valid over a remarkably wide range of adapta—
tion levels and conditions of contrast.

Color Measurement

 

"For years color matching has been an art. Now
there is a need for an orderly scientific approach to the
solution of color problems. Instrumentation could comple-
ment the practical experience of the colorist."2

Macbeth,3 in I'Color Matching," states that research-
ers are all too familiar with the discontinuities which con-
tinually arise between subjectively perceived and objective-
ly measured colors. Standardization depends on finding a
prOper balance between human and technical color evaluation.

In order to keep a prOper balance between the two, neither

aspect of colorimetry, subjective or objective, should be

 

lWright, p. 51.

2Douglas Graham, "Color Measurement and the Colorant
Mixture Computer in the Textile Industry," American Dyestuff
Report, Vol. LIII (August 3, 1964), 673-678.

3Norman Macbeth and Warren B. Reese, "Color Match-
ing," Illuminant Engineering, Vol. LIX (June 1964), 461-471.

 

ll

ignored in a comprehensive study of color. Each form of
colorimetry has specific advantages and disadvantages. The
eye is the final arbiter of color because, in the final
analysis, the color that is actually seen is most important.
However, as is continually stressed in the literature, in-
strumental measurement is also extremely important. Hunter
states that although the eye is an agile comparator, it is

a relatively poor memorizer. The eye is a relatively poor
memorizer because it contains no meters and cannot assign
numbers to the color impressions that it perceives. Instru—
mental color measurement and color specifications are neces—
sary in color research so that investigators everywhere can
communicate intelligently about important characteristics of
the appearance of color.1 Instrumentation should not
replace what the human eye can do but should complement it
with numerical and graphical expressions of sense impres-

C 2
Sions.

Objective Color Measurement

 

Wright3 has encouraged the use of spectrophotometry
by color researchers because spectrOphotometric measurement
of energy distribution does not depend on visual peculiari-

q

ties of the eXperimenter; thus the color specification is

 

lHunter, p. 9.
2Graham, pp. 673-678.

3Wright, p. 130.

12

not subject to error from this cause. Because of the great
advantages of using the Spectrophotometer, Wright feels that
this instrument will be adopted as the ideal method of color
measurement.

The spectrophotometer measures in absolute terms and
with adequate precision the energy distribution, wavelength,
reflectance, and transmission factors of lights. With many
Spectrophotometers, the tristimulus X Y Z factors, or the
amounts of red, blue, and green light present in a color,
are automatically integrated. Spectrophotometric data pro—
vide a fundamental basis for an absolute language of color.

In an article entitled "Current Practical Use of
Instrumental Color Matching,"1 Frank J. Rizzo notes the
importance of spectrophotometry in that the industrial use
of Spectrophotometers is increasing for color matching dye-
stuffs in all stages of production, for predicting fiber
blends, for measuring color degradation, and for color fast-
ness evaluations. Rizzo concludes that "all our efforts in
objective color measurement are to replace visual assess—

ment.2

 

1Frank J. Rizzo, ”Current Practical Use of Instru—
mental Color Matching in Industry," American Dye Reporter,
Vol. LII (April 13,1963), 365- 369.

2Ibid., p. 366.

13

In an experiment measuring the color of small pat-
terns on cloth, P. R. Bunkalll states that he prefers to use
a Spectrophotometer in which X Y Z readings are integrated,
because he feels that previous limitations in color measure-
ment are overcome with the application of the new technique.

Hardy2 stresses that because of the spectrophoto-
meter, experiments are repeatable with adequate precision
and provide the fundamental color language so important to
color research. Judd, like Hardy, approves the use of
spectrophotometric techniques for color measurement: "The
spectrOphotometer is the most important basic tool for the
evaluation of color of industrial products as well as for
fundamental research in colorimetry."3

Fred W. Billmeyer,4 in an article concerning
research in colored plastics, states that the complete infor-
mation given by the spectrophotometer in absolute terms is

most useful where detailed information is needed, as in

color matching.

 

1P. R. Bunkall, "Measuring the Color of a Very Small
Pattern," Society of Dyers and Colourists Journal, Vol. LXXX
(May 1964), 255-256.

2Hardy, p. 9.

3Judd and Wyszecki, p. 98.

4Fred W. Billmeyer, "Color and the Coloring of
Plastics," Color Engineering, Vol. I, No. 3 (September
1963).

 

14

Spectrophotometry was listed by the respected Inter-
national Commission on Illumination as one of the methods by
which the international standard observer was derived. The
standard observer gives the exact amounts of color mixture
necessary to reproduce any wavelength of color in the spec-

l

trum.

Recently, it was reported in the Society of Dyers

 

and Colourists Journal that the recording spectrophotometer
had been used successfully to measure the color of dyed
carpet yarns. In the paper, "Use of Instrumental Match
Prediction for Recipes for Wool-Nylon Carpet Yarn,"2 the
author eXplained that yarns colored with different dye for-
mula were mounted in 20mm internal diameter cardboard tubes
at a density of 192 yarns per tube and were cut flush with
the tube. This method of representation of the yarn gave a
reproducible surface for color measurement similar in char-
acter to the cut tufts by which carpet yarns are normally
assessed visually. The color of each mounting was measured
on the recording spectrophotometer and the resulting reflec-

tance curves and tristimulus values were used in specifica-

tion for dye recipes.

 

lWright, p. 130.

2R. Walton, "Use of Instrumental Match Prediction
for Recipes for Wool—Nylon Carpet Yarn," Society of Dyers
and Colourists Journal, Vol. LXXX (August 1964), 429-430.

 

 

15

Dorothy Nickerson, in research that concerned color
measurement in textiles, converted spectrophotometric
results into Munsell notations in order to give the results
a visual referent. She also stressed the importance of the
study of color differences based upon different sets of
measurement. "It is evident that very precise measures of
color are necessary if small color differences are ever to
be as satisfactorily studied by instruments as they are by
the eye."1

The spectrophotometer can precisely measure small
color differences, and judgments based upon such instrumen-
tal measurements appear to be somewhat more accurate than
judgments made by experienced observers, and much more
accurate than judgments made by the less eXperienced

2
observer.

Subjective Color Measurement

 

When an observer is asked to compare colors and to
evaluate color difference, there may be as many different

3 c I I 0
answers as there are observers. RecogniZing this fact is

 

lDorothy Nickerson, “The Illuminant in Textile Color
Matching," Illuminant Engineering, Vol. XLIII (January-
December 1948), 416-467.

2Hugh R. Davidson and Marvin Taylor, "Prediction of
the Color of Fiber Blends," Journal of the Optical Society
of America, Vol. LIV, No. 54 (April 1964), 250.

3

 

 

Judd and Wyszecki, p. 111.

‘ " 33—

l6

necessary in any color research; however, one should also
remember Wright's statement that if a practical system of
subjective color measurement were eventually developed, its
most widespread application would probably be found in the
realm of design.1
Keeping in mind the number of variables present when

judgments are made by a human observer, such as eye adapta- 3
tion, color contrasts, and psychological interpretation, it

would seem advisable to limit, as much as possible, the

 

Juli .

variable conditions of the environment in which the color
judgments are made. Suggestions concerning methods of pre-
senting textile samples to human observers were given in two
studies.2 The suggestions included use of a standard illu-
minant and standard angle conditions for presentation of
the samples.

Artificial light is preferred to natural daylight
because the artificial lights' irradiance and spectral dis—
tribution can be defined sufficiently, as well as can be

made independent of time.3 In 1931, the International

 

1W. D. Wright, "The Needs and Prospects of Subjec-
tive Colour Measurement," National Physical Laboratory
Symposium, No. 8 (Stationery Office, 1958).

 

2Olsen Bent Buchmann, "The Objective Measurement of
Colour Changes," Danish Institute for Textile Research, XI
(1950), 128-159.

3Y. Nayatani and S. Sysyechi, "Color of Daylight
from North Sky," Journal of the Optical Society of America,
Vol. LIII, No. 5 (May 1963), 626-629.

17

Commission on Illumination adopted a light source having a
certain energy distribution as an internal standard of illu-
mination to be used for the purpose of colorimetry. This
standard light is called illuminant "C” and is equal to the
average daylight of noon in the north sky. The source is a
tungsten lamp operated at a temperature which approximately
averages daylight, and is supplied with a suitable filter.1
The angular conditions recommended for the measurement of
color of an opaque surface are that the light should strike
the surface at a 450 angle, and that the light should be
viewed along the perpendicular to its surface.2

For visual matching, care should be taken that the
surface of the sample is in a horizontal plane and that the
standard should be close to the sample and in the same plane
of the sample. Errors in Munsell values by as much as one

whole step are possible by the inadvertent tilting of the

sample's surface.3

 

lHardy, p. 4.

2Deane Judd, "Colorimetry," National Bureau of Stan-
dards, United States Commerce Circular 478 (Washington: U.S.
Government Printing Office, 1950), p. 6.

3"The ISCC Method of Designating Colors and a
Dictionary of Color Names,” National Bureau of Standards,
United States Commerce Circular 533 (Washington: U.S.
Government Printing Office, 1955), p. 7.

18

Color and Texture

The literature was reviewed to investigate the rela-
tionship of color and texture. Theoretical material and
recent research developments concerning the effect of tex-
ture on color provide a helpful background for the under-
standing of the material being studied in this paper.

Evans defines texture as the "visible nonuniformities
in the reflectance of the surface which are obviously a phys-
ical property of the surface."1 David Katz2 described the
effect of texture surface conditions upon color as

Ersheinungsueisen, or modes of appearance. According to

 

Katz, modes of appearance are surface attributes such as
luster, film, or sheen, which integrate with the surface
color and influence the way in which a color is seen.
Recently, Frank Rizzo and Alvin O'Ramsley3 noted
that for samples where the same fiber and dyes are used, the
essential difference between the color of the two textile

samples was solely due to their surface texture. Thus,

 

lEvans, Introduction to Color, p. 122.

2David Katz, The World of Color (London: Keegan,
Paul, Trench, Trubner and Co., Ltd., 1955), p. 1.

3Frank J. Rizzo and Alvin O'Ramsley, "New Color-
Measurement Instruments for Use by the Textile Industry,"
American Dye Reporter (May 15, 1961), pp. 374-384.

 

19

these authors help establish the fact that surface texture
may help account for color change.

Warburton and Lundl also reported that color may be
modified by the textural quality of a surface.

Gross2 established that change in variation of pile
structure of a texture had a modifying effect on color. In
her limited eXperimentation, variations in the height of
pile and size of yarn proved to be causal factors in the
variations in the samples' colors.

A theory has evolved to explain why texture effects
color. Luckiesh3 states that when the ends of a nap surface
face the light, the light is able to penetrate the fabric to
a considerable depth; thus the color of the texture appears
deeper, due to multiple reflections. If, however, the nap
ends are away from the light there is more specular reflec-
tion away from the surface and less penetration of the light
into the surface. The result is less color change.

Luckiesh4 also describes the effect of certain

fibers upon color. The reflected color of a textured

 

lP. Warburton and G. V. Lund, "Colour and Textiles,"

The Journal of the Textile Institute Experiments in Colour,
Vol. 47, No. 5 (May 1956), 315-318.

 

2Gross, p. 40.

3Luckiesh, p. 309.

41bid., p. 309.

20

material is the color of the incident light after it has
been altered by the selective absorption of the fabric. In
a loose fabric with a porous surface, the light is able to
penetrate more deeply and the reflected color is the result
of multiple reflections. Wool fibers are more transparent
than cotton fibers and therefore permit the deeper penetra-
tion of the light and a greater number of multiple reflec-
tions. The light reflected from less porous fibers or fab-
rics is just slightly changed because it does not allow
multiple reflections. "This tends to dilute the colored
light and to make it appear less saturated."l

The ultimate concern in the study of textural modi-
fication of color is not whether texture modifies light and
therefore color, but rather to what extent it does and
whether predicable limits can be set as to how much texture
does affect color. In order to set predictable limits,
exact quantitative measures of color change due to textural
variation should be made. Present researchers in the area
of color and texture seem most often to describe the quali-
tative change in color that results from textural variation.
For example, in the work completed by Warburton and Lund,2

changes in hue as well as changes in value were described by

observers looking at 4% denier material, 3 denier material,

 

1Ibid., p. 308.

2Warburton and Lund, p. 315.

 

21

and finally 1% denier material. The investigators stated
that the green fabrics appeared a little more yellow and the
purple fabrics appeared bluer as the fiber Size decreased.
This researcher believes that the tendency to pre-
sent results of research in a descriptive manner rather than
to state quantitative results is not unique to the Warburton
and Lund experiment, but is, in general, the trend in re-
search concerning the modifying effect of texture upon color.
Considering the newness of this field of investigation, it
is perhaps a prudent approach for research to be presented
in such a manner. At some future time, the quantitative
results of various investigators may be compiled and then
practical limits may be set as to the extent texture does

affect color.

W

In recent years, more investigators have become
interested in studying the effect of texture upon color.
Color theory indicates that there is a physical eXplanation
for the modification of color caused by texture. Research-
ers, however, are left with the task of developing effec-
tive color experimentation that results in precise quanti-
tative and qualitative measurement of color change.

Because of the present interest in exact color

measurement, leading color researchers suggest the use of

22

instrumental color measurement to complement the work done
by human observers. The increasing effectiveness of re-
search in the area of color and texture indicates that, in
the future, predictable limits may be established concerning

the extent of the influence of texture on color.

CHAPTER II

METHODOLOGY

Research Design

 

This study was designed to describe and evaluate the
effect of three structural variables of a pile texture (pile
height, size of yarn ply, and pile density) upon the color
red-purple. The study utilized both subjective and objec-

tive methods of investigation.

Samples

Twenty-four pile samples used for analysis in this
study were made from three colors of yarn: red-purple, red,
and purple. The red-purple samples were the major concern
of this study, while the red and the purple samples were
used as controls.

Each of the colors of yarn were woven into eight
possible combinations of the three structural variables,
pile height, yarn ply, and pile density (see Table 1).

Thus, each sample with a certain pile structure
matches another sample exactly, except for the alteration
of one variable factor. For example, number one sample

matches number three sample except in pile height. Later,

23

24

 

 

 

 

 

Table 1. Research design for samples: combinations of
variables
Combination Variables
Pile (P) Density (D) Ply (PL)

1 high (HP)* high (HD)* three-ply (3PL)*
2 high (HP) low (LD)* three-ply (3PL)
3 high (HP) high (HD) one-ply (lPL)*
4 high (HP) low (LD) one-ply (lPL)
5 low (LP)* high (HD) three-ply (3PL)
6 low (LP) low (LD) three-ply (3PL)
7 low (LP) high (HD) one—ply (lPL)
8 low (LP) low (LD) one-ply (lPL)

 

*
Letter codes used throughout study to identify the
variable.

25

it will be shown that analysis of the data was done with one
variable at a time. By matching the samples into pairs, and
then studying the color differences between such pairs, any
color change between them was judged to be the result of the
one variable that made the samples different.

The pile samples were woven in the following manner:
the red-purple, red, and purple yarn used was three ply "Z"
twist construction, in which each single ply was a combina-
tion of two yarns in an "S" construction. The warp was a
heavy 5/10 medium twist, linen yarn. The weft was the pre-
viously described red-purple, red, and purple Persian rug
yarn. The pile samples were woven on a table loom and the
flossa knot technique which is the half ghiordes knot was
employed.1

The knots of all the samples were tied over one of
two height flossa bars which determined the pile heights of
one-half or one—and-one-half inches. The one-half inch pile
height was chosen to conform with the Gross study. The
variable of two pile densities was achieved by using either
two or three rows of heavy wool filler yarns between each
row of knots and by using either sixteen separated yarns i
in each knot or eighteen separated yarns (constant with the

Gross study). The two yarn sizes used were three ply and

 

lOsma Couch Gallinger and Josephine Del Reo, Rug
Weaving for Everygne (Milwaukee: Bruce Publishing’Co.,
1957).

 

 

26

one ply. The one ply yarn was a double strand of untwisted
three ply yarn. The approximate overall size of the samples,
3% X 3%", was achieved by weaving ten rows of fourteen knots
each. Because the investigator had a tendency to make
tighter knots as She progressed with the making of the pile
samples, the overall dimension of the samples varied slight-
ly. .A two inch by two inch portion in the center of each
sample was marked and this portion was analyzed by both the
human observers and the spectrophotometer. Within the two
inch demarcation, each sample contained exactly the same num-
ber of knots. This procedure assured uniformity in the sam-
ples. After the samples were woven and before they were cut
apart, they were spray coated on the back surface with a
synthetic rubber compound. Later, the top of each sample

was sheared to insure uniform pile height for each sample.

Analysis

As in all colorimetry, the fundamental question to
be answered requires the study of the stimulus (light wave-
length) and the resulting human visual response (perceived
color). This eXplains the objective and subjective ap-
proaches to color analysis used in this study. Three judges
were used for the subjective analysis portion of this study
to coincide with the three judges used in the Gross study.
However, a spectrophotometer was used in this study to help

determine whether this instrument could measure as affectively

 

27

color change due to textural variation as did the instrument
used in the Gross study. All the findings were converted
into Munselll nomenclature to facilitate easier understand-
ing of the data.

A comparison of results of this research and the
preceding Shipley and Gross studies was not made because of
the following reasons. 1

1. The Shipley pilot study was unpublished material

that was not readily available.

2. The results of the Gross study were averaged;

 

the data from the present study was not.
3. An instrument other than a spectrophotometer
was used in the Gross study to collect objective

data.

Objective Color Analysis

The General Electric Recording Photoelectric Spectro-
photometer2 was used, so that maximum information concerning
the wavelengths of the colors in the study could be found.
The General Electric Recording Photoelectric Spectrophotom—
eter, specifically designed for the measurement of spectral

reflectance and transmission, was used in this study to

 

lAlbert H. Munsell, A Color Notation (New York:
Munsell Color Co., 1919), p. 111.

2Committee on Colorimetry of the Optical Society of
America, Science of Color (New York: Thomas Y. Crowell Co.,
1953).

 

 

28

analyze objectively the color of the carpet samples. This
instrument employs a double monochromator for the elimina-
tion of stray energy. The energy supplied by this monochro-
mator passes through a polarizing photometer. An additional
polarizing prism is rotated at high speed to produce two
flickering beams which alternately attain their maximum in-
tensities. One of these beams is reflected from the sample
which was placed before the instrument through a two inch
aperature and the other beam is reflected from the standard
of magnesium oxide. The reflected light is averaged by fur-
ther reflection within a hollow Sphere lined with magnesium
oxide, after which the light finally falls upon the photo-
electric cell. The fluctuations in the photocell current
resulting from irregularities in the reflected intensities
of the flickering beam are amplified and actuates a motor
that rotates the balance prism of the photometer. Through

a mechanical coupling, the reflectance corresponding to the
condition of the optical balance is recorded on a graph
sheet, which is rotated continuously under the recording pen
as the wavelengths change. Simultaneously, the attached
tristimulus integrator automatically ascertains and records
on three dials the tristimulus values of the color sample.
Each carpet sample was presented four times to the spectro-
photometer and an average of the tristimulus values was

taken as recommended by Judd and Wyszecki.l

 

1Judd and Wyszecki, p. 100.

 

29

The tristimulus values, X Y Z, for each pile sample
were then used in a color difference formula that D. Nicker-
son and Stultz perfected in 1943.1 Before the values of
X Y Z obtained from the spectrophotometer became usable in
the formula, each value had to be converted in a Munsell
value function Vx Vy Vz. This conversion was done through
the interpolation of the data found in the appendix of Color
in BusinesngScience, and Industry by Judd and Wyszecki.2

The values Vx Vy V2 for each sample was compared

using the following formula in order to determine color dif-

ferences due to changes in pile height, yarn size, and pile

density:
[SE = 0.23 (Vy - Vyl)2 + [(Vx - Vxl)]2 + 0.4
l l 2
[(Vz-VZ)-(Vy-Vy)] 35
[AB = the unit of color difference

Vy = the Munsell value function obtained from the
tristimulus value Y.

 

 

 

Vx = the Munsell value found from the same function

of Vy setting Y = Xc.

_ x _ x

XC ‘ x MgO ‘ 98.04
Vz = Munsell value for Y = Zc

ZC=__z__:__Z_

Z MgO 118.0

1

Ibid., p. 296.

21bid., appendix.

 

30

Vyl Vxl Vzl indicate the Vy Vx Vz values for the

second sample of a pair that was compared to
first sample.
The Nickerson and Stultz equation was programmed and the

numerical data obtained gave values for 11E.

Subjective Color Analysis

The three judges who participated in the subjective Fa
color analysis were members of the faculty at Michigan State A
University. All were knowledgeable in the field of color.

The Macbeth Munsell Colorimeter Type Number One was

 

used in this part of the color analysis to help control the :9
environment in which the color assessments were made. The
colorimeter is a box-like structure with two hinged doors
on the front. The interior of the box, painted a neutral
gray (Munsell N8), is illuminated with a light source Sim-
ilar in temperature to that of North Sky daylight. The
light source is composed of two R40 300-watt Reflector Flood
lamps used with two 7%" Macbeth daylight filters. The doors
of the colorimeter were opened just far enough to eXpose a
mask bar mounted in front of the samples. The mask had two
rectangular apertures through which the observers were to
view the samples and standard. The doors helped to shield
out extraneous light.

A complete book of Munsell color chips provided the
standards for comparison. The subjective color judging

involved the following steps.

31

l. The colorimeter was made ready (doors opened,
mask in place, lights turned on).

2. The sample texture was placed inside the box on
a plastic holder which held the sample so that
the nap of the carpet was directly toward the
viewer.

3. The observer then selected a Munsell chip, and
held it next to the texture sample in the same
angle plane as the texture sample.

4. The samples were viewed in a given order: red,
purple, and red-purple, in order that the ob—
server's eyes did not become too adapted to one
color.

5. Each judge made his selection with only the
investigator present, who recorded the judge's

estimates of color.

The above stated procedure was established in the
Gross study and certain innovations were made to adapt the

procedure for this study.

Comparison with the Gross Study

Although this study stemmed directly from the Gross
study, changes in the research design and the availability
of more sophisticated instrumentation for collecting data
invalidated the possibility of directly comparing the Gross

study and this study. The factors which contributed

 

32

specifically to the dissimilarity between the two studies

were the following:

I.

II.

Subjective Color Analysis

A.

In the Gross study, the notations of the judges
were averaged.

In this study the data was not averaged.

In neither study were sufficient numbers of
color matches made to statistically negate the

possible chance of error.

Objective Color Analysis

A.

B.

The Hunter-Gardner Colorimeter was used in
the Gross study to measure color difference.
A General Electric Recording Photoelectric
Spectrophotometer was used in this study to

measure wavelengths.

 

CHAPTER III
SUBJECTIVE COLOR ANALYSIS

The color notations of the judges were analyzed in
order to ascertain whether variations in pile structure,
size of yarn ply, pile density, and pile height affected the
color of the samples.

The Effect of Variations of Yarn Plygon
Red, Red-Purple and Purple

When the color notations reported by the three
judges were analyzed for the effect of variation in yarn ply
on the color of the samples, it was found that as the size
of the yarn ply decreased, the judges perceived color
changes, on various occasions, in all of the three attri-
butes of color: hue, value, and saturation (see Table 2).

Only two instances of hue change due to variations
in yarn ply were perceived by Judge A. Both of these hue
changes were from 7.5 RP to 10.0 RP. These changes in hue
were in a clockwise movement on the Munsell color wheel.
This clockwise movement indicated that the judge perceived
the sample color as becoming more red and less purple-blue

as the yarn ply of the sample decreased.

33

ul- ~MJ...

 

34

 

 

 

 

H+ H+ oa\mmmm.s oa\mum\~ mmm.s nu ma mm
a: w\mmmm.s ma\mnoH\m mmm.s am mu mm
oa\mmmm.s ma\mnoa\m mmm.s on mm mm
ma\muoa\mmmm.s oa\m mmm.s mm mm mm
oa\m mmoa ma\muoa\m mmoa as an m

ma\muoH\m mmoa ma\muoa\~ mmoa am an m m
ma\muoH\m mmoa ¢H\muoa\m mmoa as mm m
ma\muoa\~ mmoa ~H\muoa\m mmoa mm mm m
oa\m mm.m oa\m mm.m on mu m
oa\m mm.m ma\muoa\m mm.~ om mg. m
¢H\muaa\e mm ca ¢H\¢ mm.m on mm m
6H\¢I¢H\v mm.m ¢H\¢ mm.m mm mm m
m\m mmoa m\m mmoa mg mg mm
H+ m\m mmoa m\m mam.s am an mm
m\m mmoa m\m mmoa mu mm mm
H+ m\m mmoa m\m mmm.s mm mm mm
oa\m mmoa oa\m mmoa on mg m

oa\m mmoa oa\m mmoa am an m a
H+ os\muos\m mmoa os\~ macs on mm m
oa\m mmoa oa\m mmoa mm mm m
oa\m mm.m oa\m mm.m on mu m
H+ oa\m mm.m m\m mm.m on mg m
oa\m Mm.m oa\m mm.m on mm m
m\muoa\m mm.m oa\m mm.~ mm mm m

COHDMH5umm m5am> mam mam mco ham mouse mmuduxme mamfimm MOHOU macaw
munmno Hammcsz mnu co mmm©5b mnu mnu mo
mmcmno HOHOO mo pHcD ma mGOHumuoz HOHOU Hammad: musuosuum mHHm ucmumcoo
mamasm cam .mamusm

IUmH .cmu so wan cum» mo COHDMHHM> mo uommmm mSu mCH3onm mflmhamcm HOHOU m>wuomhnsm

.m mabme

35

 

 

 

 

m\~ mam.s oa\mum\m mmm.s on ma mm
m\m mmm.s oH\mum\m mmm.n mm mu mm
H+ m\m mmm.s +m\m mmm.s on mm mm
H+ H+ oa\m mmm.s m\m mmm.s mm mm mm
Enoa\m mmoa oa\m mmoa on mg m
m+ oa\mus mmoa s oa\m mmoa om mg m o
m+ oa\~us mmoa s oa\m mmoa sq mm m
oa\m mmoa oa\m mmoa mm mm m
oa\m mm.m oa\m am.~ on mu m
oa\m mm.m oa\m mm.m om mu m
oa\m mm.m oa\m mm.m on an m
+oa\m mm.~ oa\m mm.m mm mm m
cofiumusumm msam> mum mam mco ham mouna mmusuxma mamamm Hoaou mmUSH
muumzu Hammad: mnu so mmmpsb on» may mo

mmcmso noaoo mo DHGD

mg mcoHUMDOZ HOHOU Hammad:

mnzuosuum mafia unnumcoo

 

 

 

Umsaflucoollm magma

36

A decrease in yarn ply produced a value change--the
color change observed was a one value level increase on the
Munsell chart. Judge A perceived a value change between
pairs of purple samples with a constant pile structure (HP-
LD). Judges B and C each perceived a value change between
different pairs of red-purple samples that had a constant
pile structure of (LP-LD) and (HP-HD), respectively.

Each judge perceived at least one instance of satu-
ration change due to a variation of yarn ply. Judge A per-

ceived the only saturation change that occurred in the red

‘L‘.
V

 

samples. Judge B perceived a one level increase in satura-
tion for a pair of red-purple samples with a constant pile
structure (LP-LD), and one level decrease in saturation
between another pair of red-purple samples (LP-HD). Judge
C perceived a three level increase in saturation between
pairs of purple samples with a constant pile structure
(HP-LD), and (LP-HDL respectively.

From the foregoing data dealing with the effect of
a decrease in size of yarn ply on color, it may be concluded
that whenever color changes were observed by the judges, the
changes indicate that the hue of the samples increased in
warmth, thus, becoming more red as the color moved clockwise
on the color wheel; the value of the samples increased, thus,
becoming lighter; and the saturation of the samples increased,

thus, becoming more intense.

37

The Effect of Variations of Pile Density
on Redijed-Purple, and Purple

Contradictory evidence of color change was observed
when the notations of the judges were analyzed in order to
determine the influence of a decrease in pile density on
color (see Table 3).

The only observation of hue change between a pair of .5
samples that differed from one another in pile density was
perceived by Judge A. The hue change between a pair of
red-purple samples was from 7.5 to 10, representing a clock- A
wise movement on the Munsell color wheel.

Each judge occasionally perceived value changes in
the color of the samples which varied only in pile density.
One judge perceived a one level value increase between one
pair of purple samples. Judges B and C perceived one level
decreases in value between pairs of red-purple samples which
had constant pile structures of (LP-3PL), and (HP-lPL),
respectively.

There was a lack of agreement among the judges con-
cerning the nature of the saturation change of some Samples
due to a decrease in pile density; Judge A perceived the
only saturation change in the red samples; this was a one
level increase in saturation. Judge B perceived saturation
changes only in the red-purple samples. Two pairs of red-

purple samples appeared to decrease in Saturation one level

38

 

 

 

 

annurnziizaiuwu

H: oa\mmmm.s m\m mmm.s qma as mm
H- H- oa\mum\mmmm.s ~H\m-oa\m mmm.s qmm ma mm
ofl\mmmm.s mH\muoa\m mam.s gas mm mm
ma\muofi\mmmm.n oa\m mmm.s qmm mm mm
oa\m mmoa ~H\muoa\m mmoa qma mu m

NH\~uoa\m mmoa ma\muoa\m mmoa qmm mg a m
ma\muoa\m macs ~H\muoa\m mmoa gas as m
¢H\muoa\m mmoa ma\muoa\m mmoa qmm mm m
oa\m mm.m oa\m mm.m qma mu m
oa\m mm.m NH\mu0H\m mm.~ qmm mu m
va\a mmm.m pa\¢uva\s mm.m qma mm m
aa\¢ mm.m ¢H\¢ mm.m qmm mm m
m\m mmoa m\m mmoa gas as mm
H+ m\m mmoa m\m mmm.s qmm ma mm
m\m axed m\m mmoa and mm mm
H+ m\m mmoa m\m mmm.s qmm mm mm
oa\m mmoa oa\m mmoa ASH mg m

oa\m mmoa oa\~ mmoa qmm mu m a
H+ oa\m mmoa oa\m mmoa qma mm m
oa\m mmoa oa\m mmoa. qmm mm m
oa\m mm.m oa\m mm.m gas as m
H+ oa\m mm.m m\m mm.~ 4mm an m
oa\m mm.m m\muoa\m mm.~ gas mm. m
oa\m mm.m oa\m mm.~ qmm mm m

COHDMH50mm msHm> mam mufimcmm 30a huflmcmm swam monsuxme mHmEmm uoaou mmosb
muumno Hammad: mfiu co mmmU5b msu on» mo
mmsmno HOHOU mo DHCD >9 mSOHumuoz Hoaoo Hammssz muduusuum maflm uamumcoo
mamusm cam .mamusmlpmn
.Umn CO muflmcmp mafia mo mcoflumfium> mo uommmm mflu mcHBOnm mflmmamcm HOHOU m>HuoanDm .m wanna

39

 

 

 

m\~ mam.n m\m mmm.s qma ma mm
oa\mum\m mmm.s oa\mum\m mmm.s qmm mu mm
H- oa\m mmm.s oa\m amm.s qma mm mm
+m\m mmm.s m\m mMm.s qmm mm mm
m: Snoa\~ mmoa oa\mus mmoa ASH an m
oa\m mmoa Euoa\m mmoa qmm an m o
m+ oa\mus mmoa oa\m mmoa qma mm m
sloa\m mmoa oH\m mmoa gmm mm m
oH\m mm.m oa\m mm.m sad as m
oa\m mm.m oa\m mm.m qmm mu m
oa\m mm.~ +oa\m mm.m Ana mm m
oH\m mm.m oa\m mm.~ qmm mm m
COHDMH5umm 05Hm> mdm huflmcma 30a %0Hmcmm nmflm mmuduxme mamfimm HOHOU mmCSH
muumso Hammcsz mnu so mmmvsb on» 030 m0

mmswnv HOHOO mo uHcD

>9 mcoflumuoz HOHOO Hammcsz

musuonnum mafia acoumcou

 

 

 

UmSQHDQOUIIm manma

40

as the pile density decreased. Judge C perceived an in-
crease oftflnxualevels of saturation between a pair of purple
samples with a constant pile structure of (HP-lPL), and
between another pair of purple samples with constant factors
(LP-lPL) saw a decrease of three levels.

Color changes that may be attributed to change in
pile density are not as clearly defined as were those color
changes attributed to change in yarn ply. Evidence was not
conclusive as to the dominant direction of change of the
three attributes of color for all the samples as pile den—
sity decreased. Generally, the hue of the samples remained
constant or changed in a clockwise direction on the Munsell
color wheel; the values of some samples were reported as
lighter and others were reported darker; the saturation of
most samples decreased and, thus, appeared less intense.

The fact that the judges in many instances gave
contradictory color judgments as to whether there were in-
creases or decreases in value and saturation when they noted
any changes at all indicates that color changes due to varia-
tion of pile density were difficult to discern. The diffi-
culty may be attributed to the fact that the samples did not
differ enough in pile density to make discernment of color
easy. It is also possible that pile density does not

appreciably influence color.

41

The Effect of Variations of Pile Height on
Red, Red-Purple, and Purple

 

Analysis of the subjective findings revealed that
the judges perceived more color changes when pile height
was altered than they had perceived when yarn ply or pile
density was altered (see Table 4).

A hue change of one Munsell unit was perceived
between pairs of red samples with a constant pile structure
of (LD-lPL) by Judge B. As the pile of this pair of samples

decreased, the color was perceived to move in a clockwise

 

direction from 10 RP to 2.5 R on the Munsell color wheel.
Because Judge B chose 10 RP only once to describe the color
of a red sample, this investigator felt that this judgment
may have been an error.

,All three judges observed that all the paired sam-
ples appeared darker in value as the pile height of the
samples decreased. Each value change that was indicated by
the judges was a decrease of one level on the Munsell value
scale. Judge B saw a value change between every pair of
red samples. The other two judges saw no change between
these paired samples. Only one judge, A, saw a value change
between a pair of purple Samples. Judge B and C perceived a
value change between two pairs of red-purple samples.

In all but one instance the saturation changes that

were observed by the judges were decreases in saturation.

lun- .
.. 7‘.’

42

 

 

oa\m mam.» oa\m mmm.s qua on mm
H: m\m mmm.s oa\m mmm.n qu mm mm
H: H: m\m mmm.s oa\m mmm.s Sam on mm
oa\m mmm.s oa\m mmm.s man mm mm
oa\m mmoa oa\~ mmoa gas on m
oa\m mmoa oa\~ mmoa Ama mm m m
oa\m mmoa oa\m mmoa dam on m
oa\m mmoa oa\m mmoa amm mm a
N: a: H+ oa\m mm.~ ¢a\¢ axes and an m
m- H: oa\m mm.m va\¢ mm.m gas as m
m: a: oa\m mm.~ ¢H\¢ mm.~ dam on m
m- H- ms\muoa\m mm.m ¢H\¢ mm.m Sam on m
m\m mmoa m\m mmoa Ana on mm
m\m mmoa m\m mmoa qma mm mm
m\~ mmoa m\m mmoa dam on ad
m\m mmm.s- m\m mmm.n qmm mm mm
H: oa\m macs oa\muoa\m macs qu an m
oa\m mmoa oa\~ mmoa ASH mm m
oa\m macs oa\~ mmoa Sam an m a
oa\m mmoa oa\m mmoa 4mm a: m
oa\m mm.m oa\m mm.m Ama on m
oa\m mm.m m\muoa\m mm.~ qua mm m
oa\m mm.m oa\m mm.m man as m
H- m\m mm.m oa\m mm.m qmm mm m
GOADMHmem 05Hm> mam mHHm 30A mHHm nmflm mmusuxmfi mHmEmm HOHOU mm©5b
muumno Hammad: mnu co mmmodb on» may mo

mmcmco HOHOO mo uflcb

an mGOHumDoz HOHOO Hammzsz

muduofiuum mHHm

ucmumcoo

 

 

.Umu so uflmflmn mHHm m0

mamasm 6cm

.mamusmlpmu

coflumflum> mo uummwm mnu @Gfl3onm mHmMHmcm HOHOU m>HuUan5m

.¢ magma

43

 

 

 

an m\m mmm.s oa\m mmm.s dad on mm

H- A- m\m mam.s oa\m mam.» qu mm mm
oa\mum\m mam.s +m\m mmm.s qmm on mm
oa\mum\m mmm.s m\m mmm.s qmm mm mm
m- SuoH\m mmoa oa\mus mmoa gas on m

m+ oa\m:s mmoa oa\~ mmoa qma mm m o
oa\m mmoa sloa\m mmoa dam on m
Suoa\~ macs oa\m mmoa qmm mm m
oa\m mm.m oa\m mm.m and on m
oa\m mm.m +oa\m mm.~ ASH mm m
oH\m mm.m oa\m mm.m dam on m
oa\m mm.~ oa\m mm.m 4mm mm m

SOHumusumm m5am> mum mafia 30A maflm swam mouduxms mamamm HOHOO 0065b
mphmsu Hammcsz map so mmmUSb m£u mnu mo

mmcmsu HOHOO mo Daub

ma mcoflumuoz HOHOU Hammad:

mnsuosuum mafia ucmumcou

 

 

 

Umscflucoollw magma

44

Judge C saw the only increase in saturation. This was an
increase of three levels of saturation and occurred between
a pair of purple samples. Judge B perceived that all of the
red samples decreased two saturation levels as the pile
height of those samples decreased. The color difference
between two samples with a constant structure of (HD-3PL)
appeared to Judge A as a decrease one level in saturation.
Judge C, the only judge who perceived any saturation change
in the purple Samples, indicated a decrease of three levels
of saturation between a pair of samples that had a constant
pile structure (LD-lPL).

The dominant patterns of color change due to vari-
ations in pile height were established by Judge B. This
judge perceived all of the red samples decreasing one value
level and two saturation levels as the pile height decreased.
The color judgments of the other two judges generally rein-
forced the patterns of color change established by Judge B,
although the color judgments of the other judges were made
on colors other than red. The judges reported more color
changes in the samples that varied in pile height than in

those that varied either in yarn ply or pile density.

45

The Effect of Variations of Pile
Structure Upon Red-Purple

 

 

When the red-purple color notations of judges were
analyzed, certain specific characteristics of that color
were noticed. As the pile structures of the samples were
varied, judges more frequently reported color changes be-
tween pairs of red-purple samples than were reported for the
other two colors. Also, the redwpurple samples were the
only ones perceived by the judges to change in all three
attributes of color-~hue, value and saturation. Further,
each of the judges perceived as the color he saw most fre-
quently, for the red—purple samples, a Munsell color differ-
ent from the other judges. Judge A saw 10 RP 2/8 most fre~
quently regardless of the textural variations of the red-
purple samples. Judge B saw 7.5 RP 3/10 most frequently
and Judge C saw 7.5 RP 2/8 most frequently. With the other
two colors, red and purple, all three observers generally
agreed that the color they saw most frequently regardless
of the textural variation of the samples was 2.5 R 3/10 or

10 PB 2/10 for the red and purple samples, respectively.

Summary

The subjective color analysis was done by a panel

of three judges. The judges matched the color of each

 

46

sample with a Munsell color chip within the controlled
environment of the Macbeth-Munsell Colorimeter.

The notations of the judges showed that variations
of yarn ply, pile density, and pile height did affect the
color of some samples. Of the three variables, pile height
appeared to have the greatest influence on the color of the

sample. No regular pattern of color change was found.

 

CHAPTER IV
OBJECTIVE COLOR ANALYSIS

The following is a discussion, first, of the numer-
ical expressions of color difference, (XE, found between
samples by the Nickerson and Stultz color difference formula
and, second, of an evaluation and comparison of reflectance

curves of the sample colors (see Table 5).
Color Difference Formula

The Effect of Variations of Size of Yarn
Ply on Red, Red-Purple, and Purple

For all three colors, the eXpression of color dif-
ference between pairs of samples was 0.0008 to 0.001, as the
yarn ply decreased. According to the National Bureau of
Standards’guide established in 1940,1 color difference as
small as those found would not be seen by the human eye.
Therefore, color difference resulting from a decrease in

yarn ply was negligible.

 

1F. L. Rizzo, "Quartermaster Research and Engineer—
ing Command," American Dye Report, Vol. V, No. 6 (March 20,
1961), pp. 211-220.

47

1“ .

513:3;

HP .... .

.MAV mo mcoflmmmumxm 030 umnuo
mnu Eonm coupe>mp umgp mmadfidm wo muflmm cmmBDmQ mocmanMHp uoHoo mmumoHUcH
*

 

48

 

ommao.o oomao.o .ommoo.o gas as

ommao.o osmao.o ompao.o qu om

ommao.o .mmaoo.o oosao.o qmm on

opmso.o 4mopoo.o ommao.o qmm am “asap: mafia
,mvmoo.o smaoo.o Haaoo.o qma ma

pmaoo.o 4mvooo.o mmaoo.o qmm ma

memoo.o .mpooo.o mpmoo.o qma mm

Ramoo.o mmmoo.o Homoo.o amm mm shamamp mafia
mmmoo.o mamoo.o psmoo.o on ma
.maooo.o HHHoo.o ammoo.o am an
wwwooo.o ammoo.o mmmoo.o an an

masoo.o amsoo.o smsoo.o mm mm was cums
madman mamusmlpmm pom mmusuxme mamemm magmaum>

AMAV v MHSEHom NDHSum + COmmeUHZ may %9
pmHSmmmE wam>flumuflucm50 mUGmHmMMHQ HOHOU

may no

mmsuosuum maam unnumcoo

 

 

unmnmn mafia no
mNHm moguflm CH Hmfluocm mco Eouw HmMMHG umnu mmamfimm ammzumn mocmummwflc HOHOU

.huflmcmp mHHm

.mam cumm mo

.m OHQMB

49

The Effect of Variations of Pile Density
on Red, Red-Purple, and Purple

The numerical expressions of color differences tabu-
lated by the computer for pairs of samples that varied from
one another in pile density ranged from 0.0006 to 0.001.

The [XE for many of the samples clustered from 0.004 through
0.002.

Pile density produced greater color differences than
changes in yarn ply. The reader should be reminded that
these eXpressions of color difference are quantitative in
nature, not qualitative. From the data, it is known that
color differences between samples do exist but it remains
unknown whether these color differences are in hue, value,
or saturation.

The Effect of Variations of Pile Height
on Red, Red-Purple, and Purple

Color differences recorded for samples that varied
in pile height were far greater than differences recorded
for samples that varied in either size of yarn ply or pile
density. The range of color difference between high and low
pile samples was 0.008 and 0.01. Most of the color differ-
ences were recorded at 0.02.

The data, as a result, indicate that of these three
variables (pile height, yarn ply, and pile density) pile

height had the greatest relative effect upon the color of

50

the samples. The color difference between most of the pairs
of texture samples was 0.02. This is not a perceptible
color difference for the human eye since the National Bureau
of Standard guide defines a color difference of 0.0 through
0.5 as‘a “trace" of color difference.

Even though the color differences that were found to
exist between samples as pile height, yarn ply, or pile
density was altered were too small to be called a "trace"
by NBS standards, these color changes were measureable by
the spectrophotometer. This indicates to this researcher
that perhaps greater textural changes would have created
greater color differences.

The Effect of Variations of Pile Structure

Upon the Color Red-Purple

Further examination of the data indicated in Table 5
showed that on seven occasions (starred on the charts) large
discrepancies existed between the color differences obtained
for matched pairs of one color samples as compared with
pairs of matched samples of the other two colors.

Earlier the Nickerson and Stultz formula was used to
discover how much color difference existed between sample
textures which varied from one another in only one character-
istic. All factors of color and pile structure were kept
constant except for one characteristic of pile structure,

such as yarn ply, pile density, or pile height. Therefore,

 

51

this researcher believes that each given variable of pile
structure should affect the same amount of color difference
between pairs of samples even though these pairs of samples
are different-—red, red—purple, or purple. For example, it
may be seen that a variation in the size of yarn ply was
responsible for a color difference of 0.00787, 0.00787 and
0.00715 between pairs of red, red-purple and purple samples
with constant pile structures of (HP-HD). The closeness of
these numerical expressions of color differences should be
expected if no factor other than yarn ply was affecting the
color of the Samples.

The data, however, indicate that upon seven occa-
sions the numerical expression of color difference between
matched samples of one color were unlike the eXpressions of
color difference between pairs of samples of the other two
colors. Four out of seven times it was the color difference
between red-purple samples that was dissimilar. On all four
occasions the color difference between the red-purple sam-
ples was far less than the color difference between pairs of
red or purple samples. The dissimilarity indicates to this
researcher that perhaps it was some innate quality of the
colors themselves that influenced the amount of color change

exhibited by the samples as pile structure was varied.

52

The Effect of Variations of Pile Structure
on Red, Red-Purple, and Purple

Reflectance Curves

 

The spectrOphotometric data were also used to eval-
uate the influence of variations of pile structure on the
reflectance curves for each individual sample texture. The
color difference formula provided a quantitative measure of
color difference between two compared samples which varied
from one another in one factor of pile structure only.
Examination of the reflectance curves of those same pairs of
samples provided the qualitative measure of color difference
between two samples. Such an examination of the color
reflectance curves of matched samples indicated that as the
texture of each sample changed, certain definite changes
took place in the color of each sample (see Table 6).

The size of yarn ply variable had little or no
effect on the amount and kind of color reflected from the
red-purple and purple samples. The pairs of reflectance
curves of the red samples, however, demonstrated that de-
creasing the size of yarn ply resulted in loss of reflec-
tance in the 600 to 700 millimicron portion of the spectrum.
This meant that as the yarn ply size decreased the sample
became less red (a loss in saturation) and became perhaps a
little darker, because a loss in percentage reflectance

indicates a darker color.

53

1.. .n ‘ Olv

.mmHSEMm SuHmsmp 30H mo mmHH mmmusmoumm m mmumoHUCHa

 

 

 

 

 

SE oon OH SE och Wm SE 005 wh
SE omalomfi N SE omvloow m ...... ... HSH SH
SE oon mo SE oon HH SE oon NH
SE ova 6 SE omwloow m SE owsloow m HSH Hm
SE oon m SE 005 Wm SE 005 S
SE mm¢ ma SE onwloo¢ wH SE onwloow N HSm SH
as cos 3 as con m as con NH 233
SE owwlomw wm SE oovloov H SE on¢loo¢ m HSm Hm mHHS
SE 005 m ... ... SE 005 ms
... ... ... ... SE ovwloo¢ Ha HSH SH
... ... SE oon H SE oow mH
ooo coo co. co. SE OQfilOOfl w qmm mg
SE omo mm SE oon m SE oou m.
SE oww wm ... ... SE ovaloow H HSH Sm
as co» 4 as co» m: as con m simamp
SE ova H ... ... SE oovnoos H HSm Sm mHHS
SE ovo m SE oon wH SE oonlooo S
SE mma wH ... ... SE osmloww mH SH SH
SE omo H SE oow w SE ochlovo m
SE mma H ... ... SE ovalooa H mm SH
... ... SE oos H SE oonlomo m
as m2. H . . . . . . as omauooa w SH Sm
... ... SE oon .mm SE 005 To S S
SE oomlooa H ... ... SE oovuoow mH SH SH H Sum»
HumSmHm>m3 m SMHO
x S» m0 soHSHOS x
mHSuSS mHSHSSIUmS pmm mmuSuxmB mHSEmm mnu mo mHHmHHm>
mnumSmHm>m3 mHSuUSHum mHHS usmumsoo

:Hmuumo um HoHoo comm HOS mm>HSO musmuumHmmS mo
mHHMS SmoBumm mocwumMMHQ mo mmmucmonS SH SOHO

 

 

mHSHSS cam .mHSHSSIUmH .Umu co mHSuoSHum mHHS mo mSOHuwHHm> mo
uommmm mnu mo mm>HSU mosmuomHmmu mo somHHMSEoo msH3OHm mHmSHmsm MOHoo m>HuumflHO .w mHHmE

54

As the samples' pile density decreased, the result—
ing color change between pairs of samples apparently was
extremely slight at both ends of the spectrum. The red
samples lost reflectance at both the blue and red ends of
the Spectrum. As the pile density was decreased, the red-
purple samples' color changed the least.

Variation in pile height resulted in relatively
large color change in Samples of all three colors. Great
losses in reflectance were common among samples of all three
colors at about 600-700 millimicrons. Two pairs of red sam-
ples demonstrated the greatest individual reflected percent-
age loss in the red region of the spectrum. The purple sam-
ples lost considerably more reflectance at 430-450 millimi-
crons than did the other two colors.

The reflectance curves indicated that decreasing
pile height produced a decrease in the amount of red
reflected. Because the reflectance curves of the low pile
samples dropped at both ends of the spectrum 400, and 700
millimicrons, a darkening of the sample was indicated.

The reflectance curves for the red—purple samples
showed that as size of yarn ply was decreased and pile den-
sity was decreased, the percentage reflectance loss of
various wavelengths for that color was more similar to per-
centage reflectance loss for purple than for red. The per-
centage reflectance loss for each color was more significant,

however, when pile height was reduced. The loss in

 

55

reflectance of the red—purple samples was far more similar
to the loss suffered by the red samples than to the loss
among the purple samples.

When comparing actual reflectance curves, those of
the red-purple were also similar to the red. Consequently,
both colors reflect nearly the same type and quality of

light and have a similar spectral nature.

Summary

The objective color analysis was done by the General

Electric Recording Spectrophotometer. The data were analyzed

 

in two ways: (1) the Nickerson and Stultz color difference
formula, and (2) comparison of the reflectance curves.

The Nickerson and Stultz formula showed the color
difference between samples to be extremely small. Pile
height was the one variable that influenced the greatest
amount of color change between samples.

Comparison of the reflectance curves revealed that,
generally, as pile structure varied there was a decrease in
the percentage of reflectance in the red region of the spec-
trum. Red and red-purple lost the greatest single amounts
of red reflectance. Red—purple reflectance curves were

similar to the red reflectance curves.

CHAPTER V

COMPARISON OF THE SUBJECTIVE AND OBJECTIVE

METHODS OF COLOR ANALYSIS

In comparing the spectrophotometric tristimulus data,
translated into Munsell nomenclature, with the responses
given by the three judges (see Table 7), a great deal of
agreement between both methods of color analysis was dis-
cerned. Generally, the spectrophotometric color analysis
and the judge's color perceptions for a given sample indi-
cated the same Munsell color, 2.5 R 3/8 through 2.5 R 3/10,
7.5 RP 3/8 through 7.5 RP 3/10, and 10 PB 2/8 through 10 PB
2/10.

There were occasions when the spectrophotometric
results did not agree with the color notations of the judges.
This investigator feels that the reason for the lack of spec—
trophotometric reinforcement of the subjective notations was
primarily because the color differences between pairs of sam-
ples varying only in one structural detail (size of yarn ply,
pile density, and pile height), was very small indeed. The
spectrophotometer was able to analyze the smallest trace of
color change between samples of varying structure as is evi-
dent by the extremely small values of color difference calcu-

1ated with the Nickerson and Stultz formula.

56

57

 

 

m\~ SSm.s OH\m SSm.s m\~ SSOH 0H\mum\m SSS.» HSH SH SH SS
m\m SSm.s m\m SSm.s m\m SSOH OH\mum\m SSm.s HSH SS SH SS
m\m SSm.s m\~ SSm.a m\~ SSOH OH\mum\m SSm.s HSm DH SH SS
m\~ SSm.s OH\m SSm.» m\m SSm.S OH\mum\m SSm.S HSm SS SH SS
OH\N SSm.s OH\m SSm.s m\~ SSOH OH\mum\m SSm.S HSH SH SS SS
oH\m SSm.s OH\m SSm.s m\~ SSOH OH\mum\m SSm.s HSH SS SS SS
+m\m SSm.s OH\m SSm.s m\~ SSOH OH\muw\m SSm.s HSm SH SS SS
m\m SSm.a OH\m SSm.s m\m SSm.s OH\mum\m SSm.s HSm SS SS SS
OH\N SSOH OH\N SSOH OH\N SSOH OH\mum\m SSOH HSH SH SH S
.xms SSoH OH\N SSoH 0H\~ SSOH m\m SSoH HSH SS SH S
OH\N SSOH OH\N SSOH 0H\m SSOH m\m SSOH HSm SH SH S
OH\N SSOH OH\N SSOH OH\N SSOH 0H\mum\m SSOH HSm SS SH S
.xms SSOH 0H\m SSOH 0H\m SSOH OH\mum\m SSOH HSH SH SS S
0H\m SSOH OH\N SSOH OH\N SSOH QH\mum\m SSOH HSH SS SS S
OH\N SSOH OH\N SSOH oH\m SSOH OH\mum\m SSOH HSm SH SS S
OH\N SSOH OH\N SSOH 0H\m SSOH OH\Num\m SSOH HSm SS SS S
OH\m Sm.~ OH\m Sm.m OH\m Sm.m OH\mum\m Sm.m HSH SH SH S
OH\m Sm.~ OH\m Sm.~ OH\m Sm.m OH\mum\m Sm.m HSH SS SH S
QH\m Sm.m OH\m Sm.m OH\m Sm.m OH\mum\m Sm.m HSm SH SH S
OH\m Sm.m 0H\m Sm.m m\m Sm.m oH\mum\m Sm.m HSm SS SH S
0H\m Sm.m ¢H\a SSOH OH\m Sm.~ OH\mum\m Sm.m HSH SH SS S
+OH\m Sm.m vH\a Sm.m IOH\m Sm.m 0H\mum\m Sm.~ HSH SS SS S
0H\m Sm.~ aH\a Sm.m 0H\m Sm.m 0H\mum\m Sm.m HSm SH SS S
QH\m Sm.m ¢H\a Sm.~ OH\m Sm.~ OH\mum\m Sm.m HSm SS SS S
O m d SmumEouosS mHSEmm HoHoo
mmmUSb IouuomSm.

 

 

mHSumHUSmEOS HHmmSSE OSSH UmumHmSmHu

SmumEouozmouuomSm SSS HSHB mmmpSn mmunu mnu mo mmmcommmu map mo SOmHHmSEOU

.m SHAME

58

The three judges undoubtedly sensed the same small
color differences. They could not quantitatively report
these differences; the spectrophotometer, however, could
report exactly how close the match of the color of the sam—
ple texture was to the color of the Munsell chip. However,
when the judges thought that the color of a sample was some-
what brighter than the closest matched Munsell chip, they
would be forced to use the next closest Munsell chip which
was actually far brighter than the color of the sample
texture. There were two reasons for such action by the
judges. First, all the judges were aware of the purpose of
this research and were therefore mentally set to perceive a
difference in color due to a difference in texture. Second,
the judges were hampered by the non-linear relationship1 of
the Munsell chip. Because of this, they were unable to indi-
cate where the color of the samples fit on the differing
lengths of the color continuum between one Munsell chip and
another. The tristimulus values provided by the spectro-
photometer indicated quickly how close the color of the sam-
ples matched the tristimulus specification for each Munsell
chip. In this way, it became apparent to this investigator
that in nearly all cases, the tristimulus specifications for

both the samples and the Munsell chips did not match exactly,

 

lThe Munsell Color System is a visual system and
therefore the spacing between color chips varies in length.

59

but the specifications were not off enough to justify giving
the color of the sample another Munsell notation. Therefore
the spectrophotometer did not indicate any color change
between samples of varying structure because the small color
differences that were manifested in the color difference
formula were too minute to be indicated in a change in
Munsell notation as the three judges apparently tried to do.

Not only was there no spectrophotometric reinforcem
ment of the color differences perceived by the judges, but
also, at no time did any of the judges reinforce each other's
indication of color change. Therefore, the conclusions con-
cerning trends in color change caused by variations in pile
structure were sometimes established on the basis of one
indication of color change made by one judge. Also the
number of color matches made by the judges were not numerous
enough to negate statistically the possibility of chance
selection and error. For the reasons stated above, this
investigator feels that in this study, the data received
from the spectrophotometer concerning the colors of the sam~
ples are more valuable than the subjective color analyses
because these data indicate far more accurately the quanti-
tative and qualitative change in color resulting from alter-
ation of pile structure of the samples.

In the previous discussion, the primary concern was
to compare the quantitative color change recorded by both

subjective and objective methods of color analysis. Further

6O

analysis of the data indicated that the results of both the
subjective and objective methods of color analysis agreed
more with one another with respect to the qualitative color
change than quantitative color change that took place as the
pile structures were varied.

The subjective color analysis showed that as pile
density and pile height were decreased in the samples, the
value levels of the colors of the samples decreased as did
their saturation levels. Analysis of the reflectance curves
by the spectrophotometer of each Sample reinforced the
judges' findings.

The objective analysis of the color of the samples
showed that the greatest change in color always occurred in
the red region of the spectrum. The judges reported most of
the color changes in the red-purple paired samples with the
second greatest number of color changes in the red samples.
This, plus the fact that the reflectance curves of all the
red-purple samples were more similar to the reflectance
curves of the red samples than were the purple samples has
all contributed to the hypothesis that perhaps it was the
influence of red in the color red-purple that was responsi-
ble for the difficulty with which the color red-purple was

discerned in this study and in the previous study.

 

CHAPTER VI

SUMMARY AND SUGGESTIONS FOR

FURTHER RESEARCH

Summary
In 1961, Gross completed a study on ”The Effect of
the Height and Yarn Size of a Pile Texture on Color." In

this study, the author reported that texture affects color
in a manner measurable by both objective and subjective
techniques.

The results of the Gross research showed that while
the majority of the colors studied changed in a consistent
direction as the texture of the samples was altered, a few
samples, particularly, the blue-green, yellow-green, and
red-purple, exhibited change that deviated from the rest.
Gross hypothesized that the hand dying of the blue-green and
yelIow-green yarns was the cause of irregular changes. The
author suggested further research in order to understand
better the unique behavior of the color red—purple when
modified by texture.

This present study was designed to study the effect
of variations of pile structure upon the color red-purple.

For this purpose, three sets of samples were woven that

61

62

varied from one another in pile structure. One set of sam-
ples was woven with red yarns, one set with purple, and
with red-purple yarn. The pile structures of the samples
in each set varied in size of yarn ply, pile density, and
pile height. The combination of variables resulted in a
total of 24 samples.

A subjective color analysis was made by a panel of
three judges. The judges matched the color of the sample
with a Munsell chip within the controlled environment of the
Macbeth-Munsell Colorimeter. In addition, color analysis of
each sample was also made on a General Electric Recording
Spectrophotometer.

The color notation for each sample was compared with
the notation of every other sample so that the compared
pairs of samples differed one from the other in only one fac-

tor, size of yarn ply, pile density, or pile height.

Summary
I. As the yarn ply of the samples decreased the results
were as follows:
A. Subjective Analysis
1. The hues of the samples became warmer.
2. The value of the colors increased.
3. The saturation of the colors increased.
B. Objective Analysis
1. Color difference between samples was from 0.008

to 0.001 NBS unit of color difference.

63

2. Reflectance curves showed that the pairs of red
samples especially differed from one another in
the percentage of reflectance in the 600-700

millimicron section of the spectrum.

II. AS pile density of the samples decreased the results
were as follows:
A. Subjective Analysis
1. The hues of the samples became warmer.
2. The values of the samples decreased.
3. The saturation of the samples decreased.
B. Objective Analysis
1. Color differences between pairs of samples were
found to be from 0.006 to 0.001 NBS unit of
color difference.
2. Reflectance curves showed that between pairs of
samples there was a loss in percentage reflected,
mostly in the red region but also in the blue

region of the spectrum for all three colors.

III. As the pile height of the samples decreased the results
were as follows:
A. Subjective Analysis
1. The hues of the samples remained unchanged.
2. The values of the colors of the samples decreased.
3. The saturation of the colors of the samples

decreased.

64

Objective Analysis

1.

Color difference between samples was found to be
from 0.008 to 0.01 NBS unit of color difference.
Reflectance curves showed that between pairs of

samples there was a large loss in percentage of

‘reflectance in the 600-700 millimicron region of

the spectrum for all three colors. Purple sam-
ples also lost reflectance in the 400-480 milli-

micron region of the spectrum.

IV. The data revealed the following specific characteristic

of red-purple.

A.

Subjective Analysis

1.

The judges perceived more frequent numbers of
color change between pairs of red-purple samples.
As pile variations were made, red-purple was the
only color to change in all three attributes of

color.

Objective Analysis

1.

On four occasions, the [5E for pairs of red-
purple samples deviated from the [SE for red and
purple.

As variations in pile structure were made, the
percentage of reflectance lost in the red region
of the spectrum for red-purple was similar to

red.

3.

65

Red-purple reflected the same type and quality

of light as the red.

From the above findings certain general conclusions

were made:

1.

Changes in the pile structure of the sample
textures were not great enough to permit easy
discernment of color change.

The spectrophotometer was able to analyze and
record the slightest color difference between
samples.

Pile height was the variable in pile structure
that affected the greatest color change.

The difficulty with which the color red-purple
was discerned was due to the fluctuating red
portion of that color.

Red-purple showed a greater pattern of irregular

color change than did either red or purple.

Recommendations

This study has presented many interesting challenges

to this researcher. The task of measuring the effect of

texture on color requires an understanding of many fields of

knowledge.

To explore this topic further, the scientific

implications of the physical aspects of light as they per-

tain to the effect of texture on color phenomena should be

66

more thoroughly investigated. Moreover greater variations
in pile texture, eSpecially in pile density, might result in
more definitive answers. If subjective color analysis is
used, a larger group of judges should make repeated color
matches at different times, so that the possibility of

eXperimental error can be reduced.

BIBLIOGRAPHY

Books

Committee on Colorimetry, Optical Society of America.
Science of Color. New York: Thomas Y. Crowell Co.,
1953.

 

Crewdson, Frederick M. Color in Decoration and Design.
Illinois: Frederick J. Drake and Co., 1953.

Gallinger, Osma Couch and Josephine Del Reo. Rug Weaving
for Everyone. Milwaukee: Bruce Publishing Co.,

Graves, Maitland. Color Fundamentals. New York: McGraw-
Hill Book Co. Inc., 1952.

Hardy, A. C. Handbook on Colorimetry. Cambridge: Tech-
nology Press, 1936.

Judd, Deane and Gunter Wyzsecki. Color in Science, Business
and Industry. New York: John Wiley and Sons, Inc.,
1963.

Katz, David. The World of Color. London: Keegan, Paul,
Trench, Trubner and Co., Ltd., 1935.

Koblo, Martin. World of Color. trans. Ian F. Finlay.
New York: McGraw-Hill Co. Inc., 1956.

Luckiesh, Mathew. Light, Vision and Seeing. New York:
D. Van Nostrand Co. Inc., 1944.

Munsell, Albert H. A Color Notation. New York: Munsell
Color Co., 1919.

Wright, W. D. The Measurement of Colour. London: Helger
and Watts, 1958.

67

68

Articles and Periodicals

Billmeyer, Fred C. "Color and the Coloring of Plastics,"
Color Engineering, XLIII (January-December 1948),
416-467.

Buchmann, Olsen Bent. "The Objective Measurement of Colour
Changes," Danish Institute for Textile Research, XI
(1950), 128-159.

Bunkall, P. R. "Measuring the Color of a Very Small Pattern,"
Society of Dyers and Colourists, LXXX (May 1964),
255-256.

Davidson, Hugh R. and Marvin Taylor. "Prediction of the
Color of Fiber Blends," Journal of the Optical
Sociepy of America, LIV, (April 1964), 250.

Evans, Ralph M. "Variables of Perceived Color," Journal of
the Optical Society of America, LIV, No. 12 (December
1964), 1467-1474.

Graham, Douglas. "Color Measurement and the Colorant Mix-
ture Computer in the Textile Industry," American
Dyestuff Report, III (August 3, 1964), 673-678.

Hunter, Richard S. "The Other Aspects of Appearance,"
Color Engineering, I, No. 2 (June 1963), 8-14.

Judd, Deane B. "Colorimetry," National Bureau of Standards,
Circular 478 (Washington: United States Government
Printing Office, 1950).

Macbeth, Norman and Warren B. Reese. "Color Matching,"
Illuminant Engineering, LIX (June 1964), 461-471.

Nayatani, Y. and S. Syszechi. "Color of Daylight from
North Sky," Journal of the Optical Society of
America, LIII, No. 5 (May 1963), 626-629.

Nickerson, Dorothy. "The Illuminant in Textile Color
Matching," Illuminant Engineering, XLIII (January-
December 1948), 416-467.

Rizzo, Frank J. and Alvin O'Ramsley. "New Color Measurement
Instruments for Use by the Textile Industry,"
American Dye Reporter (May 15, 1961), 374-384.

69

Rizzo, Frank J. "Current Practical Use of Instrumental
Color Matching in Industry," American Dye Rpporter,
LII (April 13, 1963), 365-369.

Rizzo, Frank J. "The ISCC Method of Designating Colors and
Dictionary of Color Names," National Bureau of
Standards, Circular 533 (Washington: United States
Government Printing Office, 1955).

 

Walton, R. "Use of Instrumental Match Prediction for
Recipes for Wool-Nylon Carpet Yarn,“ Society of
Dyers and Colourists Journal, LXXX (August 1964),
429-430.

Warburton, P. and G. V. Lund, "Colour and Textiles," The
Journal of the Textile Institute EXperiments in
Colour, XL, No. 5 (May 1956), 315-318.

Wright, W. D. "The Needs and Prospects of Subjective Colour
Measurement," National Physical Laboratory Symposium
No. 8 (London: Stationery Office, 1958).

Unpublished Material

Gross, L. "The Effect of Height and Yarn Size of a Pile
Texture," Unpublished Master's thesis, College of
Home Economics, Michigan State University, 1961.