. ' '. ..,' ’ _ o .‘--‘ .1 . . "0' _ . I COSTUME SILHOUETTES AND FASHEOM IDEALS 0F BEAUTY: - . 21840T01940' mm forthgflegree of m. A. . woman 37m umvmsm ELAINE LEE PEDERSEN ' 1975 I .v- ‘ c - ‘ 9:: - l . . - - - 3‘31”. . . . ‘ . 1 l'. I .0. ‘. . . h ' . . . L“ 'Q‘klAJNR... I. ‘A - Q ‘i ' - n q -‘. .‘ V‘ a , THESIS 331333331333333 1113333 ""‘" “”1”" ‘S'flzz’ny; ‘.' .0: 3M: ' I. .~ - “.9 £9.09 ' T- L! B R A. * f-“f'chigan 3 -, ‘ 1L: ~--.w:rsu/ 5M £3.” g“? amomg’av "‘ 3333 3 3338' i 3333313033vmc. 3 ”as F—x 097/ 5". ’ [V ‘.' . .1 1| . J I ABSTRACT COSTUME SILHOUETTES AND FASHION IDEALS OF BEAUTY, 1840 TO 1940 BY Elaine Lee Pedersen The purpose of this study was to investigate the fashion ideal of beauty--that concept of feminine beauty illustrated in fashion magazines and promoted by the clothing industry. The fashion ideal of beauty portrays the newly introduced characteristics of beauty and/or dress. This study examined differences in the fashion ideal of beauty in body pr0portions and frequencies of types of shoulder slopes, facial shapes and hairstyles between magazines; changes in the fashion ideal of beauty over time; and whether differences existed between the Greek standard of beauty and the fashion ideal of beauty. There were five major hypotheses. 1. During a costume silhouette period the fashion ideal of beauty will not differ between randomly selected issues. Elaine Lee Pedersen 2. During a period of transition of costume silhouettes the fashion ideal of beauty will not differ between randomly selected issues of two fashion magazines. 3. The fashion ideal of beauty found in randomly selected issues of two fashion magazines will not change during a defined costume period. 4. The fashion ideal of beauty found in randomly selected issues of two fashion magazines will change during a costume transition period. 5. There will be a difference between the body prOportions of the Greek standard of beauty and the mean body prOportions of the fashion ideal of beauty from 1840 to 1940. The data were randomly selected from Godey's Lady's Book, Peterson's Magazine, Harper's Bazaar, and Vogue although only two magazines were used at any one period. The time covered by the study, 1840 to 1940, was divided into costume periods and costume transition periods based on Agatha Brooks Young's research of skirt silhouette periods.1 A measure was develOped from previously used methods of measuring clothing illustrations that could be applied to measuring the fashion figure. A system similar to that illustrated in clothing selection books was chosen where the unit of measure would be headlengths. The following body proportions were measured: total height, torso length, waist to floor length, shoulder width, and Elaine Lee Pedersen waist width. Types of shoulder 510pes, facial shapes, and hairstyles were tabulated. None of the five hypotheses was completely rejected. No difference in body proportions was found between magazines for two of the costume periods and one costume transition period. During costume periods no change was found in the mean prOportions of the fashion ideal of beauty for certain groups of years and for the frequencies of some of the characteristics of the fashion ideal of beauty. Change was found in the fashion ideal of beauty during costume transition periods for certain years and characteristics. The most conclusive finding was that, except for shoulder width from 1862 to 1872 and waist width from 1905 to 1940, there was a difference in every period between the Greek standard of beauty and the fashion ideal of beauty. lAgatha Brooks Young, Recurring Cycles of Fashion (New York: C00per Square Publishers, Inc., 1966), p. 20. COSTUME SILHOUETTES AND FASHION IDEALS OF BEAUTY, 1840 TO 1940 BY Elaine Lee Pedersen A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Human Environment and Design 1975 ACKNOWLEDGMENTS I should like to express my appreciation to Dr. Anna Creekmore, my thesis advisor, for the hours of con- sultations and recommendations during this study; and to Dr. Holly Schrank and Dr. Sadayoshi Omoto, my committee members, for their advice and guidance. I would also like to thank Dr. Robert R. Rice for his support as well as Dr. Kathleen Musa, who was an original member of my committee and who continued to advise me throughout my study. Recognition is due to Nancy Martin for her assistance in the statistical analyses and for setting up the computer program for the analyses. Special thanks go to my family for their assistance and support; my grandmother, Mrs. Luette Handy; my mother, Mrs. Nita Pedersen and my sister, Ms. Marilyn Pedersen. Thanks also go to Kathy Cyr, Lois Florkey, and to Sara Butler for their suggestions and help during this study as well as to Martha Brownscombe and Terri Zensen for their moral support and encouragement. ii TABLE OF CONTENTS Page LIST OF TABLES O O I O O O O O O O O O 0 v Chapter I. INTRODUCTION . . . . . . . . . . . II. REVIEW OF LITERATURE. . . . . . . . . 3 Ideals of Beauty . . . . . . . . . 3 Fashion Ideal of Beauty . . . . . . . 5 The Meaning of Fashion . . . . . . . 9 Costume Periods. . . . . . . . . . 15 Summary . . . . . . . . . . . . 17 III. STATEMENT OF THE PROBLEM . . . . . . . 19 Definition of Terms . . . . . . . . 19 Assumptions . . . . . . . . . . . 21 Hypotheses . . . . . . . . . . . 23 IV I PROCEDURE 0 O O O O O O O O O O 0 26 Selection of the Sample . . . . . . . 26 Development of the Method of Measuring of Fashion Ideals of Beauty . . . . . . 30 Collection of the Data . . . . . . . 34 Statistical Analyses . . . . . . . . 35 V. RESULTS AND DISCUSSION . . . . . . . . 36 Description of the Sample . . . 37 Costume Period: Difference in the Fashion Ideal of Beauty Between Magazines . . . 40 Costume Transition Period: Difference in the Fashion Ideal of Beauty Between Magazines . . . . . . . . . . . 47 iii Chapter Costume Period: Change in the Fashion Ideal of Beauty . . . . . . . . . . . Costume Transition Period: Change in the Fashion Ideal of Beauty . . . . . . Difference Between the Classic Greek Figure and the Fashion Ideal of Beauty . Summary . . . . . . . . . . . . VI. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS. . Summary . . . . . . . . . . . . Conclusions . . . . . . . . . . . Implications. . . . . . . . . . . Recommendations. . . . . . . . . . BIBLIOGMPHY O O O O O O O O O O O O O O APPENDICES Appendix A. Method of Measuring Illustrations . . . . B. Selection of Data. . . . . . . . . . iv Page 51 57 6O 63 66 66 67 70 71 75 80 89 LIST OF TABLES Table Page 1. Diagram of Time Periods and Magazines Selected to Test the Hypotheses . . . . . . . . 24 2. Selection of Fashion Magazines . . . . . . 27 3. Sample Breakdown. . . . . . . . . . . 39 4. Range of Values Found for Body PrOportions by Costume and Transition Periods . . . . . 41 5. Differences Between Magazines, Costume Periods, Multivariate Analyses . . . . . . . . 43 6. Differences Between Magazines, Costume Periods, Chi Square . . . . . . . . . . . . 45 7. Differences Between Magazines, Transition Periods, Multivariate Analyses . . . . . 48 8. Differences Between Magazines, Transition Periods, Chi Square . . . . . . . . . 50 9. Change in the Fashion Ideal of Beauty Within Costume Periods, Multivariate Analyses. . . 52 10. Change in the Fashion Ideal of Beauty Within Costume Periods, Chi Square . . . . . . 55 11. Change in the Fashion Ideal of Beauty Within Transition Periods, Multivariate Analyses. . 58 12. Change in the Fashion Ideal of Beauty Within Transition Periods, Chi Square . . . . . 59 13. Differences Between the Fashion Ideal of Beauty and the Greek Standard of Beauty, One Way Analysis of Variance . . . . . . 61 CHAPTER I INTRODUCTION ‘f'. V E‘Eomparisons of fashion magazines seem to indicate that there may be a difference in the portrayal of fashion illustrations over the years, although at any One time the manner of portrayal of model types seem similar. Authors of fashion literature discuss costume silhouette changes but few of them mention the similarities or differences in the portrayals of styles. There does exist literature describing body prOportions of ancient Greek statues which are used by students studying figure drawing techniques.‘ Contemporary clothing selection books also illustrate the Greek body prOportions as a standard against which student's figures can be evaluated. Students are taught how to manipulate their clothing in order to appear to have the standard proportions. The prOportions found in Greek statues do not always appear to be the same as the figure types illustrated in the fashion magazines. JThe portrayals of models in the magazines tend to show exaggerations of different parts of the body and/or costume from time to time. Perhaps an ideal of beauty inSpired by fashion and promoted by the magazine editors and fashion illustrators exists. If so could such a "fashion ideal of beauty" change over time according to the styles of the period? The purpose of this study was to investigate the concept of a fashion ideal of beauty as portrayed in the fashion illustrations in relation to the changes in the I costume silhouette. The concept of the fashion ideal of beauty as portrayed at any one time will be compared with the figure proportions of the Greek statues. By studying the fashion ideal of beauty from one costume silhouette period to the next and by comparing it with the figure pr0portions of the Greek statues it should be possible to define the fashion ideal of beauty in more concrete terms than has been done by previous investigators and to discuss the differences and similarities which occur over time. CHAPTER II REVIEW OF LITERATURE The amount of literature that discusses the concept of an ideal or a standard of beauty is limited. The following paragraphs consist of a review of the available literature on (1) ideals of beauty, (2) concepts of a fashion ideal of beauty, (3) meanings of fashion, and (4) costume periods. Ideals of Beauty 'According to Roach and Eicher, a cultural ideal of v...“ beauty of the human form represents a consensus of what a 7 t I society considers the most beautiful.1 {Aéross societies differences in cultural ideals are depicted in the vari- ations of the carved or sketched representations of human figures.2,xWestern man has in fact had some type of defined standards of figure prOportions since the time of the ancient Egyptian civilization. According to Hartley, the 4 1Mary Ellen Roach and Joanne B. Eicher, The Visible Self (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1973), p. 93. 21bid. Egyptians originated a scale of body proportions which was later perfected by the Greeks and depicted in their statues.1 From time to time artists of the Western world have studied and worked with different body prOportions although the Greek proportions for human figures seem to have been the most frequently used. Since the nineteenth century they have served almost exclusively as a standard for students of art. Many of this century's clothing selection books utilize the Greek figure as the ideal, with students being instructed to manipulate clothing design lines to achieve the illusion of the standard. Buttrick, in her book Principles of Clothing Selection, published in the 19205, referred to the "standard human figure" in relation to fashion changes: Throughout all absurdities and abnormalities that fashion has decreed at different times, an ideal persists as the basis of clothing design. This ideal is what we call the standard human figure. 1Jonathan Scott Hartley, Anatomy in Art (New York: Styles and Cash, 1891), p. 93. 2Helen Goodrich Buttrick, Principles of Clothing Selection (New York: The Macmillan Co., 1925), p. 55. More recent clothing selection texts by Morton,1 Hillhouse,2 and Chambers and Moulton3 referred to the Greek standard figure as the criteria for deciding a student's figure type and proportions. These authorities used the system of dividing the figure into headlengths for analyzing body pr0portions in order to arrive at the type of garment lines and forms needed to create the illusion of the standard. VLFashion Ideal of Beauty Although clothing selection books illustrate the Greek standard of beauty,Zfashion magazines may illustrate a different idea1--the fashion ideal of beauty. Fashion .. .gr magazines evolved from the practice of including fashion plates (which were deve10ped to promote fashion) into women's magazines. Tortora, in a history of the development and purpose of fashion magazines, stated that their major purpose was the dissemination of fashion information.4 1Grace Morton, The Arts of Costume and Personal A earance, revised by Mary E. Guthrie, Villettaféite, and June Ericson (New York: John Wiley and Sons, Inc., 1964), pp. 31-420 2Marion 8. Hillhouse, Dress Selection and Design (New York: The Macmillan Company, 1963), p. 10. 3Helen G. Chambers and Verna Moulton, Clothing Selection (Philadelphia: J. B. Lippincott Company, 1969), p. 49. 4Phyllis G. Tortora, "The Evolution of the American Fashion Magazine as Exemplified in Selected Fashion Journals 1830-1969" (unpublished Ph.D. dissertation, New York University, 1973), p. 3. Young,1 Kroeber,2 and other investigators have used fashion magazines as a source of information on newly presented clothing styles. " Various authorities have referred to the illustrated fashion figure as related to drawing techniques or illus- trative technology. There is no literature on the concept of a portrayed ideal of beauty in fashion magazines. {/xButtrick, however, discussed the idea of a shared concept of figure type in terms of fashion change in this way: "The basis of fashion lies in varying the pr0portions of the silhouette and in placing the emphasis upon different parts of the body at different times;£§” She believed that fashion change resulted when the pr0portions of the silhouette varied and when the emphasis shifted to differ- ent parts of the body.4 Based on her analysis of fashion change and that of clothing selection authorities there may be, in fact, two different ideals of feminine prOpor- tions. One ideal would be an established standard of beauty with prOportions that would not change over time. lAgatha Brooks Young, Recurring Cycles of Fashion (New York: Cooper Square Publishers, Inc., 19667: pp. 37-38. 2A. L. Kroeber, "On the Principle of Order in Civilization as Exemplified by Changes of Fashion," American Anthropologist, XXI (July, 1919), p. 289. 3Buttrick, Principles of Clothing Selection, p. 55. 4Ibid. The proportions of the second ideal would be dictated by fashion and thus would change with clothing style changes. Flugel1 and Laver2 (who showed evidence of a familiarity with Flugel's works) were concerned with the origins and use of clothing and the possible motivations behind the initial use. Both authors discussed changes in clothing styles from a Freudian vieWpoint and concluded that the change occurred because of necessary shifts in the emphasis on particular erogenous zones of the body. This explanation of changing body emphasis created through changing styles supports the concept of an ideal of beauty inspired by fashion. Another ideal, the "feminine ideal," was discussed by Anspach. She defined the feminine ideal which "each 'girl strives for," as young and glamorous, beautiful and attractive to men.3 This feminine image could also be .r / considered an ideal of beauty.‘/Afispach stated that fashion personifies the prevailing ideal of beauty (ideal of femininity) because fashion represents what is potentially ,3 acceptable to the public.€W’Fashion, according to Anspach, \l‘ 1J. C. Flugel, The Psychology of Clothes (London: Hogarth Press LTD, 1930), pp. 156-62. 2James Laver, Clothes (New York: Horizon Press, 1953), p. xxi. 3Karlyne Anspach, The Why of Fashion (Ames, Iowa: The Iowa State University Press, 1967), p. 31. 41bid. functions to prepare peOple for change.1 Possibly, then, the current ideal of beauty might, along with fashion, change with the public's taste,” This approach does not take Flugel's Freudian viewpoint but is similar to Cunnington's views on dress and the current ideal of beauty. Cunnington, in his book Why Women Wear Clothes, stated that the dress of the time reveals the current ideal of beauty. "Old fashion plates instruct us how the art of costume seizes on whatever physical feature is admired and emphasizes it, adding to or subtracting from the height and shape of the wearer according to popular requirements; expanding here or contracting there as required."2 There- fore, what is fashionable, according to Cunnington, reflects the current figure ideal, or that body feature which is being emphasized or idealized at the time. The fashion plate, he stated, is an abstraction of the current ideal of beauty.3 Another writer, Bigelow, also referred to the "fashion image." In describing early illustrations she characterized them in terms of the image they portrayed, for example, ". . . facial expressions were full of lIbid., p. 39. 2C. Willett Cunnington, Whnyomen Wear Clothes (London: Faber and Faber, Ltd., 1941). p. 97. 31bid., pp. 97-98. innocent charm."l This description reveals that more than costume is being illustrated; an ideal is also being portrayed in the image. This "fashion image" varies with time. For each different fashion image there is, depending on the body areas being emphasized, a different exaggeration or figure distortion portrayed by the illustrator.2 Based on the literature it appears that changes in costume styles may be accompanied by changes in a preferred figure type. The current ideal of fashion as sketched by the illustrator and portrayed in the fashion magazines may vary with clothing style changes and with changes in the portrayed figure type. The Meaning of Fashion Fashion has been discussed by authorities from several fields of knowledge. There are differing definitions of fashion among sociologists, historians, economists, and other professionals. Sociologists tend to view fashion, as well as the arts and literature, as reflective of social and political events. Costume historians often treat clothing and fashion as being synonymous and discuss "fashion" in chronologically ordered periods that may or may not coincide with style cycles. Economists 1Marybelle S. Bigelow, Fashion in History; Apparel in the Western World (Minneapolis: Burgess Publishing Com- pany, 1970), p.‘2921 2Ibid., p. 300. 10 view fashion as a commodity--merchandise to be sold--and usually refer to dress in general rather than the fashion— able items of the season. The problem in any discussion of fashion is to differentiate the what, where, when, how, and why of fashion. Discussions of the fashionable deal with the characteristics of costumes that have won recent acceptance by a few or will likely be accepted shortly. Definitions of fashion often disregard the constraints of the level of acceptance or fashionableness but focus on the movement of identifiable characteristics through introduction, wide- spread acceptance, and decline. This movement of styles is defined as the fashion process. Agatha Brooks Young Defined fashion as the newly accepted styles of this year which differ from last year's styles as they will also differ from next year's styles.1 The diffusion of the particular style features of dress over succeeding seasons along with the introduction of new ones is recognized as the process of change. The fashion- able, however, are the newly accepted styles. Because of the multiple characteristics of a costume it is possible to view fashion in terms of its parts. The parts include the silhouette and individual garment features, i.e., the sleeve or the collar. Costume silhouettes generally remain quite similar over relatively long periods of time, but lYoung, Recurring Cycles of Fashion, p. 6. ll costume features change fairly frequently, modifying the silhouette slightly while retaining the recognizable form. In general, each new fashionable silhouette differs from the past and the future silhouette. Anspach, in her book The Why of Fashion, discussed fashion in three ways, as a social phenomenon, as a commodity, and as a symbol. Although Anspach stated that the term fashion is used in many ways she was concerned primarily with how and why fashion changes.l Fashion as a social phenomenon is a process of change with wave-like cycles of successive styles that rise, reach a peak of acceptance, and then decline.2 She contended that fashion could be defined ". . . as the accepted way of talking, walking, eating, and dressing that is adopted by a group of pe0p1e at a given time."3 In this definition she seemed to be describing the mode, or the most p0pular form among a group of people, although in contrast she also stated that it was the ". . . new element that makes the new fashion . . .," and thus appeared to be differentiating fashion as widely accepted from "new" fashion. In addition Anspach also commented that fashion viewed as a social phenomenon ". . . has no objective truth . . ." because when it is finally accepted by the majority it begins to fade and is no longer fashion.4 1Anspach, The Why of Fashion, p. xi. 2 3 4 Ibid. Ibid. Ibid., p. xiii. 12 When discussing fashion as a commodity, Anspach defined it as "the prevailing style" or "the accepted way,"1 which is in reality modal dress. Fashion when defined as a commodity is thought of as a tangible object with characteristics that can be cepied, bought, and sold. Women's clothing is often Spoken of as fashion merchandise. Fashion as a symbol was defined, according to Anspach, as a form or commodity that over time has had symbolic meanings attached to it. It ". . . becomes a collective symbol of social identity . . ."2 in that items of apparel are transformed ". . . into symbols of personal 3 attributes and goals, social patterns and striving." Here Anspach appeared to equate fashion with dress in general and to discuss the symbolic functioning of clothing in social groups. .Apparel used a sufficient length of time to acquire symbolic meaning would probably be "out of fashion." Kelley discussed fashion as a rational and structured phenomenon that ". . . is an integral part of 4 the social order." She did not define fashion but instead analysed fashion theories. She reported that the theories lIbid., pp. xiii—xiv. 2. . 3. Ibld., p. x1v. Ibld., p. 259. 4 Carole Louise Kelley, "An Analysis of Two Theories of Fashion Change During the Years 1927-1934" (unpublished M.S. thesis, Cornell University, 1970), p. 18. l3 explaining the Operation of the phenomenon could be grouped into cyclical theories and social deterministic theories.l Cyclical theories were of several types: (1) cycles of product acceptance, (2) cycles of characteristics of the accepted product styles, and (3) cycles in skirt silhou- ettes. The cycle of product acceptance refers to the business aspect of clothing and is thought of as distri- bution curves of clothing product sales; while the cycle of characteristics of the accepted product styles is the movement from acceptance to rejection of individual parts of garments, i.e., collar, hem length, etc. The cycles in skirt silhouettes included only the change in skirt sil- houettes and did not differ conceptually from cycles of accepted product styles. Kelley felt that the theories of fashion that could be classified as socially deterministic emphasized a rationale for change; therefore, the concept of change of garment characteristics was fundamental to the social determinists' definitions of fashion.2 Nystrom was classified by Kelley as a social 3 determinist; he defined fashion as the prevailing style. The accepted fashion is influenced by dominating events, lIbid., p. 2. 21bid. 3Paul H. Nystrom, Economics of Fashion (New York: The Ronald Press Co., 1928), p. 4. l4 ideals, and social groups; and styles in fashion move in a cycle of introduction, acceptance, and rejection.1 Nystrom felt that the movement was more rapid from acceptance to obsolescence with increased wealth, prOSperity and leisure time, widespread education, and effective but cheap reproduction of goods. The cycle can be slowed by the opposite forces.2 Flugel might also be classified as a social deter- minist although he does not seem to fit easily into either of Kelley's categories. While Flugel did not differentiate between fashion as newly accepted styles or modal dress he seemed to be discussing the latter or commonly accepted dress.3 As discussed earlier he felt sexual feelings influenced clothing trends.4 Based on literature it appears that social deter- ministic theories are interpretative and cyclical theories are descriptive. Authorities holding both views share the belief that fashion change is a process whereby the newly accepted or fashionable styles move to greater acceptance and become less fashionable over time. Descriptive fashion theories, i.e., cyclical, have been useful for studying change in clothing silhouettes while interpretative, i.e., social deterministic, have served to explain change. lIbid., p. 83. 21bid., pp. 24-26. 3Flugel, The Psychology of Clothes, pp. 137-45. 41bid., pp. 156—60. 15 Since the purpose of the study was to measure change in garment styles or in figure prOportions influenced by garment styles the results of cyclical studies were studied for a basic structure for the investigation. Costume Periods Costume historians have generally classified costume periods according to the span of time from the introduction of a new silhouette through minor changes in style features affecting the silhouette to drastic changes of form. In 1919 A. L. Kroeber, a sociologist, investigated changes in women's evening dresses shown in fashion magazines and found that similar silhouettes occurred in specific periods of time.1 A later study by Agatha Brooks Young investigated changes in daytime dress silhouettes.2 The cycles resulting from the two studies did not coincide probably because different types of garments were studied and different methods of measuring illustrations were used in each study. Young analyzed characteristics of daytime dress finding mean garment styles for each year; Kroeber determined width and height ratios of full front views of evening dress and then determined silhouette periods from the yearly ratios. 1Kroeber, "Changes of Fashion," pp. 257-60. 2Young, Recurring Cycles of Fashion, pp. 10-11. l6 Payne,1 Laver,2 Braun-Ronsdorf,3 and Kohler,4 well known costume historians, did not measure costume or sil- houette periods but analyzed styles in relation to centuries and decades of centuries. A study of their work indicated that they generally concurred with Young's findings: the tubular skirt silhouette-1796 to 1829; bell skirt silhouette- 1830 to 1867; backfullness-1868 to 1899; and the tubular silhouette-1900 to 1937 (the termination of Young's study). Payne5 and Kohler6 differed from Young on the date of the change from the tubular to the bell silhouette. Both Payne and Kohler stated that the period of the bell skirt silhouette began in the 18203, ten years earlier than Young's findings. Payne's discussion of costume did not continue into the twentieth century but Laver, Braun- Ronsdorf and Kohler agreed with Young as to the date of the period of transition from backfullness to tubular. lBlanche Payne, History of Costume (New York: Harper and Row, Publishers, Inc., 1965). 2James Laver, The Concise History of Costume and Fashion (New York: Harry N. Abrams, Inc.71 3Margarete Braun—Ronsdorf, Mirror of Fashion, translated by Oliver Coburn (New York: McGraw-Hill Book Company, 1964). 4Carl Kohler, A History of Costume (New York: Dover Publications, Inc., 1963). 5Payne, History of Costume, p. 485. 6Kohler, A History of Costume, p. 249. 17 Young stated that the years surrounding the change from one skirt silhouette to the next were years of transition with no sharp break between acceptable silhou- ettes. During these periods the skirt silhouette of the earlier period would begin to take on characteristics of the next skirt silhouette period.1 The existence of the periods of transition would explain.Payne's and Kohler's differences of Opinion concerning the year of change from one skirt silhouette period to the next. The conclusions from Young's study were fairly con- sistent with the opinions of other costume historians as to when skirt silhouette changes occurred. Since her con- clusions were the result of a measure of fashion illustra- tions rather than a survey of historic costume as illustrated in paintings or museum collections it seemed logical that her study would be more reliable and a better basis for defining skirt silhouette periods. Usually investigators conducting surveys of historic costume do not count and analyze their findings. Their conclusions, therefore, may be less accurate. Summary According to the authors of the literature reviewed there exists a standard of beauty based on the Greek statue prOportions which is used by students participating in 1Young, Recurring Cycles of Fashion, p. 22. 18 figure studies as well as clothing selection. This particular standard of beauty has been used by students at least since the turn of the century.l ’ («'IThe concept of an ideal of beauty influenced by the currently fashionable silhouette as inspired by fashion emphasizes a particular body area. This ideal was expressed by illustrators from the fashion world. During a particular period there may be a different exaggeration of body features portrayed by the fashion illustrators than during another period. Change in dress silhouette was mentioned as occurring coincidently with the changes in body emphasis. The fashion ideal of beauty should be investigated, and its relationships to costume silhouettes should be ascertainedx//Kx At the present time there are only theoretical and vague references to the concept of a fashion ideal of beauty. A study could provide the framework for an empirical dis- cussion of the fashion ideal of beauty. lButtrick, Principles of Clothing Selection, p. 55. CHAPTER III STATEMENT OF THE PROBLEM The purpose of this study was to investigate differences in the fashion ideal of beauty between fashion magazines/during costume silhouette periods and during transition periods of the costume silhouette; change in the fashion ideal of beauty during costume silhouette periods and during transition periods of the costume silhouette; and whether differences existed between the Greek standard of beauty and the fashion ideal of beauty during any costume silhouette period or period of transition of the costume silhouette. k///Definition of Terms \/ Dress: Clothing and adornment worn by an individual. b/Style: A garment which retains one or more charac— teristics which identify it as a particular style in any fashion period but which can be modified to suit the spirit of the times. 19 20 u"The mode: The prevailing and accepted garment styles worn by most peOple in any period. y Fashion: The newly presented ideas for each costume season that are the result of a process of slow continuous change within a silhouette period. The styles of this year differ from last year's styles as this year's styles will differ from the styles of next year.1 “Costume period: A time period in which garment styles share a common silhouette. V/Ideal of beauty: A concept of beauty for any cultural group; it includes the figure type, posture, hair- style, body proportions, and facial shape and features. V/Standard of beauty: A basis of comparison, a model of the approved and accepted type of beauty. It includes the whole figure. ‘/Fashion ideal of beauty: A concept of beauty, often promoted by those in the fashion industry. The ideal portrays the newly introduced characteristic of beauty and/or dress. The fashion ideal of beauty undergoes slow continuous change. \/‘Illustration: A technique of dealing with a subject in a non-literary manner for mass production in a periodi- cal. 1Young, Recurring Cycles of Fashion, p. 6. 21 Fashion magazine: Those periodicals covering pri— marily clothing style tOpics and mainly featuring fashion illustrations. Body proportions: The proportions defined as body proportions for this study were total height, torso length, waist to floor length, shoulder width, and waist width. Shoulder lepes: The types of shoulder slopes counted for this study were square, natural and sloping. Facial shapes: The types of facial shapes counted for this study were oval, round and angular. Hairstyles: The types of hairstyles counted for this study were short, medium, long, worn on tOp of the head, and worn on tOp of the head with curls hanging down. Assumptions V1.1 Artistic ideals of female figure types can be 1 9’measured. Both A. Gladstone Jackson's and Arthur Thompson's2 works discuss female figure proportions based on the ancient Greek figure ideals as portrayed in Greek sculpture. h/2.§’Illustrations of fashion magazines are a suitable It" source of past eras' concepts of beautyi] 1A. Gladstone Jackson, The Right Way to Human Fi ure Drawing and Anatomy (New York: EmersoniBooks, Inc., 1932), pp. 9-10. 2Arthur Thompson, Handbook of Anatomy for Art Students (Oxford: The Clarendon Press, n.d.), p. 4272 22 Henry C. Pitz stated that periodicals by means of their illustrations as well as their articles influence the looks of men and women. In his article, "Millions of Pictures," Pitz cited Charles Dana Gibson as an example of the power of an illustrator over peoples' concept of beauty.1 Young felt that "Fashion magazines cannot at any time stray far from popular taste and still have any I O C 2 real chance of surv1v1ng in popular favor." 1 V3.5 Fashion silhouettes change over_time;] As a result of her study Young found that fashion silhouettes are continually undergoing change and felt that, "The fundamental principle that is always valid is that fashion changes are continuing processes or trends."3 b4II/Generally fashion magazines do not portray un- lfashionable or unmodal dress. Tortora analyzed "format, editorial and literary content, subject matter and form of illustration, proportion and variety of fashion and nonfashion subjects and the changes which occurred in the five journals she studied. She classified fashion magazines as those 1Henry C. Pitz, "Millions of Pictures," American Artist (November-December, 1961), pp. 37, 87. 2Young, Recurring Cycles of Fashion, p. 37. 3Ibid., p. 147. 23 . . . that consistently devoted close to half of their illustrative material to fashions or related topics . . ."1 ‘v5.i:Portrayal of fashion and figure types has been defined by available illustrative technology] Bigelow, in Fashion in History, Apparel in the Western World, discussed how early fashion illustrators were limited in their results by the available printing techniques.2 Hypotheses There were five major hypotheses which were investigated. Following the formal presentation of the hypotheses is a diagram by time period and source which shows the comparisons made of the data for each hypothesis (see Table 1). \J¥(Y/During a costume silhouette periodlthe fashion ideal of beauty as seen in selected full front 3 fashion illustrations found in randomly selected / issues of two fashion magazines will not differ r l. f between magazines.f full front fashion illustrations found in randomly selected issues of two fashion magazines will not 14 7 LL, I.{ The fashion ideal of beauty as seen in selected 1 1 g 1 1Tortora, "The Evolution of the American Fashion Magazine," p. 4. 2Marybelle S. Bigelow, Fashion in History, Apparel in the Western World, pp. 292-93. 24 Table l.--Diagram of Time Periods and Magazines Selected to Test the Hypotheses. Costume Transition Hypothesis Period Period Magazine I. Differences A. 1842 to 1861 Godey's Between Magazines Peterson's B. 1873 to 1875 Bazaar Peterson's C. 1905 to 1940 Bazaar Vogue II. Differences A. 1862 to 1869 Godey's Between Magazines Peterson's B. 1870 to 1872 Bazaar Peterson's C. 1894 to 1904 Bazaar Vogue III. No Change in A. 1840 to 1861 Godey's Fashion Ideal Peterson's B. 1873 to 1875 Bazaar Peterson's C. 1905 to 1940 Bazaar V0228 IV. Change in A. 1862 to 1869 Godey's Fashion Peterson's Ideal B. 1870 to 1872 Bazaar Peterson's C. 1894 to 1904 Bazaar Vogue 25 1differ between magazines during a period of t ! n i-transition of costume silhouettes. {Eyf. “During a costume silhouette period the fashion ideal of beauty as seen in selected full front fashion illustrations found in randomly selected issues of two fashion magazines will not change uring the costume silhouette period. hfl4 The fashion ideal of beauty as seen in selected full front fashion illustrations found in randomly / selected issues of two fashion magazines will i change during a period of transition of costume L——silhouettes. “g, There will be a difference between the body propor- L tions of the Greek standard of beauty and the mean body prOportions of the fashion ideal of beauty from 1840 to 1940. CHAPTER IV PROCEDURE Included in this chapter are discussions of the selection of the sample, the development of the measure, the collection of data, and the analysis of the data. Selection of the Sample By definition the fashion ideal of beauty at any time can be found in fashion illustrations. The population from which the sample was selected was originally to be a sample of the full front illustrations found in Godey's Lady' Lady's Book, 1840 to 1869; Peterson's Magazine, 1842 to 1875; Harper's Bazaar, 1870 to 1940; Delineator, 1875 to 1895; and 29322, 1895 to 1940 (see Table 2). The magazine issues selected were chosen according to Tortora's analysis of the percentage of illustrations in the above magazines that were devoted to fashion for the years 1830 to 1969.1 The Delineator was not available through inter-library loan services and was therefore not used. 1Tortora, "The Evolution of the American Fashion Magazine," p. 268. 26 27 Table 2.--Selection of Fashion Magazines. Time Period Years Magazines Costume Transition Costume Transition Costume 1840 1842 1862 1870 1873 1876 1894 1895 1905 tO to to to to to to to 1841 1861 1869 1872 1875 1893 1904 1940 Godey's Godey's Peterson's Godey's Peterson's Peterson's Bazaar Peterson's 3193;. m Delineator* Bazaar Delineator* Bazaar Vogue Bazaar Vogue *Not available. 28 Agatha Brooks Young, in her study of costume silhouettes, found that fashion changes could be accurately studied, recorded and measured. Based on her results she was able to define the following costume silhouette peridds: tubular, 1796-1829; bell, 1830-1867; backfullness, 1869-1899; and tubular, 1900-1937 (the end of her study).1 Young felt that costume silhouettes changed gradually and that some years of transition from the old style to the new existed between each costume silhouette period.2 Young's costume silhouette periods were used as costume periods for this study. The investigator felt that a ten year period would be adequate to measure the gradual changes which might occur during a transition period from one clothing silhouette period to the next. Therefore, five years prior to the year of the costume silhouette change as determined by Young and five years following that year were defined as the years of clothing silhouette transition for this study. The years were divided as follows: 1. 1840 to 1862, bell silhouette period 2. 1862 to 1872, transition 3. 1873 to 1893, backfullness silhouette period 4. 1894 to 1904, transition 5. 1905 to 1940, tubular silhouette period 1Young, Recurring Cycles of Fashion, p. 20. 2Ibid., p. 22. 29 The magazine issues used for this study were selected in two ways since all the magazine issues could not be obtained for every year of the study (1840 to 1940) a random sample was selected for the costume periods. If a magazine issue that had been randomly selected was not available the following issue was used. The total number of magazine issues randomly selected for each costume period and per each magazine was equal to the number of years in that particular period. For example from 1840 to 1861, there were twenty-two randomly selected issues, per magazine, from which data was taken. For the years of transition, because change in the fashion ideal of beauty was hypothesized, it was necessary to measure illustrations from each year of transition. Therefore, for each year of transition and for each fashion magazine one issue was randomly selected. When available, the first five full front illustra— tions were measured in each magazine issue selected. This number of illustrations was judged to be adequate and realistic both in terms of the cost of photoc0pying, time u spent in measuring and in terms of adequacy for statistical analysis. ”Individual illustrations were used in order of appearance in each randomly selected magazine issue. A full front illustration of the figure facing forward with little or no twist to the body or head was needed in order to measure the figure from side to side} 30 Development of the Method of Measuring of Fashion Ideals of Beauty The concept of the fashion ideal of beauty is not, as has been mentioned, a new concept, but it has never been clearly defined or measured. A method was developed by studying and analyzing measures that had been used by Kroeber,1 Kroeber and Richardson,2 Joseph,3 Hackler,4 and Malory.5 The methods used by Kroeber, Kroeber and Richardson, and Joseph could be most easily used because they could be adapted to body measurements. Kroeber was one of the first researchers who measured clothing styles. He was interested in measuring the change in the styles of evening dress from 1844 to 1919 1Kggeber, "On the Principle of Order in Civilization as Exemplified by Changes of Fashion," pp. 239-40. 2Jane Richardson and A. L. Kroeber, "Three Centuries of Women's Dress Fashions, A Quantitative Analysis," Anthropological Records, Vol. 5, No. 2 (1940), p. 112. 3Marjory Lockwood Joseph, "Changes in Women's Day- time Dress as Related to Other Selected Cultural Factors During the First and the Third Decades of the Twentieth Century" (unpublished Ph.D. dissertation, Pennsylvania State University, 1962), pp. 86-87. 4Ruth Nadine Hackler, "Fashion Cycles in Style, Fabric, and Design of Women's Skirts in Four Fashion Magazines from 1925-1961" (unpublished M.S. thesis, Oklahoma State University, 1962), pp. 29-33. 5Mary Ann Malory, "The Relationship Between Fluctuations in Hemlines and Stock Market Averages From 1921 to 1971" (unpublished M.A. thesis, University of Tennessee, 1971), pp. 12-15. 31 as portrayed in fashion illustrations. He measured eight positions on the illustrations, four vertical and four horizontal. These included the length of the figure, the length of the garment, the placement of the waistline, the depth of the bodice front, the width at the skirt hem, the maximum width of the skirt above the hem, the width of the waist and the width of the shoulders. He used full front figures for his study since this pose made it possible to use all of the above positions. To measure each of the eight positions he used the length of the figure as the base measure; thus each measurement of a position would be a ratio of the length of the figure.1 A study using a modification of Kroeber's method which related change in daytime dress to other cultural patterns in the first and third decade of the twentieth century was completed by Joseph in 1960. She refined Kroeber's measure and added an analysis of garment charac- teristics. Joseph concluded that the two analyses combined made a more thorough measure of garment style changes than did Kroeber's method used alone.2 1Kroeber, "On the Principle of Order in Civilization as Exemplified by Changes of Fashion," pp. 239-42. 2Joseph, "Changes in Women's Daytime Dress," pp. 82-870 32 Kroeber and Richardson used a modification of Kroeber's original method, as did Malory.1 Hackler tabulated styles of skirt, skirt designs and fabrics, number of fashion pages per magazine and magazine circu- lation figures.2 In many clothing selection books the standard ideal of beauty is measured in terms of head lengths. In the past this type of assessment had been used for unclothed figures but the system seemed feasible for clothed figures as well. (:After reviewing measures and studying illustrations in fashionmmagazines of various time periods thefollowing characteristics were chosen to be measured: total height, torso length, waist to floor length, waiSt width, garment Shoulder—width; figure shoulder width, and neck length.' ‘Not alI of the qualities of the fashion ideal of beauty are \llile-~ - - 11_ measurable in an objective way.f The fashion ideal 0 ~14.-- “4““.— ‘mggauty is the concept of beauty which includes not only the current clothing styles and figure type but also the complexion, facial structure and the current facial expressiépjwhether it be one of innocent charm or that of 1’ . . . a sexual vamp. ,Fac1al expre551on and complex1on were not considered in this study37 1Richardson and Kroeber, "Three Centuries of Women's Dress Fashion," p. 112; Malory, "The Relationship Between Fluctuations in Hemlines and Stock Market Averages," pp. 12-15. 2Hackler, "Fashion Cycles of Women's Skirts," pp. 29-330 33 Shoulder slopes, facial shapes and hairstyles could not be measured in the same way as above but were tallied by frequency of occurrence into types. The following characteristics of each were counted: 1. Hair style--short, medium, long, worn on tOp of the head, and worn on tOp of the head with some hair hanging down. 2. Facial shape--ova1, round or angular. 3. Shoulder lepe--square, natural or sloping. The angle of the shoulder slope was initially determined with a protractor. The following degrees defined the type of slope: square 0-16, natural 17-32, and sloping 33+. These categories were chosen by randomly selecting and measuring illustrations from several time periods to determine the largest angle or the most slope to the shoulder. The largest angle measured was 48 degrees which was divided by three to form the three shoulder slope types as mentioned above. Measurements were taken from the left side of the figure unless this side was not shown in the illustration and then the right side was used. After measuring a few illustrations it became clear to the investigator that it was not possible to measure both the garment shoulder width and the figure shoulder width in all fashion periods. This particular measurement was then changed to shoulder width. (See Appendix A for the method of measuring the illustrations.) 34 Collection of the Data Since the majority of the magazine issues needed for this study were not available in the Michigan State University Library, 95 percent of the illustrations used were photocopied through the inter-library loan services. Some were photocopied by the Newman Library in Chicago, and a few were traced. The measurements were made by drawing the necessary lines and marks on the c0pies or tracing paper (see Appendix A). To measure each illustration the headlength was found by measuring the distance from the tip of the chin to the center of the eyes, this length is equal to one-half a headlength and was added from the center of the eyes upward to determine the t0p of the figure. The total height was then measured, followed by torso length, waist to floor length, shoulder width, waist width, necklength, and tabulations of the type of shoulder lepe, facial shape and hairstyle (see Appendix A). After measuring illustrations from various time periods the investigator found that with the exception of one or two illustrations the measurement for the neck length would not be useful because the majority of the illustrations had a neck length of one-third of a head- length. This measurement was then discontinued. 35 Statistical Analyses The data was coded for computer use and statistical computer packages were utilized to simplify the analyses of the data. fi/ffTests of multivariate analysis of variance were 61441, b. used to test the difference between the means of the body proportions between time periods and between magazines. ,1 1x (4 Chi square contingency tables were used to determine if differences between magazines existed for types of shoulder slopes, facial shapes, and hairstyles.; . -, I. )A univariate analysis of variance was used to test V the difference between the means of the body proportions of ,n .- CHAPTER V RESULTS AND DISCUSSION The body proportions of selected figures found in a random sample of fashion magazines from 1840 to 1940 were recorded. Multivariate analyses of variances were used to test the difference between the means of the body prOpor- tions for selected time periods and between different fashion magazines. Chi square contingency tables were used to compare the differences between magazines for types of shoulder lepes, facial shapes and hairstyles. The data from the sample is described and the results from the hypotheses testing discussed as follows: (1) differences in the fashion ideal of beauty between fashion magazines for costume periods, (2) differences in the fashion ideal of beauty between fashion magazines for transition periods, (3) changes in the fashion ideal of beauty for costume periods, (4) changes in the fashion ideal of beauty for transition periods, and (5) differences between the fashion ideal of beauty and the Greek standard of beauty. 36 37 Description of the Sample The sample consisted of full front fashion illustrations selected from randomly selected issues of Godeyfs, Peterson's, Bazaar, and Vogue. The first five full front illustrations (not including advertising) in each issue were selected. The magazine issues were randomly selected for each year of transition of the costume silhouette and randomly selected for each entire costume silhouette period. Random selection was used to insure an equal chance of selecting the magazine issues; however, this type of sampling later caused problems in the analysis of the data when there were not selections from every year. A better plan would have been to use a stratified random sample for each year of the costume silhouette periods and to have made alternate arrangements in the years where magazine issues might not have been available. When five full front figures were not available in 7the issue selected, the next issue was used. The type of garment being modeled was not used as a criterion for selection nor were the possible minor differences in the stance of the model or differences in illustrative tech- niques. In six issues five full front figures were not available, and it was not possible to obtain the next issue, therefore four or less illustrations were used, as listed, for the following issues: 38 Peterson's, Volume 1, Number 2, February 1842; 3 Illustrations Peterson's, Volume 2, Number 4, October 1842; 3 illustrations Peterson's, Volume 2, Number 6, December 1842; 2 illustrations Peterson's, Volume 9, Number 5, May 1846; 3 illustrations Peterson's, Volume 64, Number 2, August 1873; 4 illustrations Vogue, Volume 33, Number 24, June 1909; 4 illustrations (See Appendix B for a complete listing of the magazine issues that were selected.) Out of a total of 834 illustrations in the original sample 789 were used in the study. Some of the measure- ments could not be taken on the remaining 45, because of the stance of the model. Table 3 shows the number of illustrations measured per costume period, transition period, and magazine. The following body proportions were measured and characteristics were noted to determine the fashion ideal of beauty: total height torso length waist to floor shoulder width waist width shoulder slope--square, natural, or sloping 39 Table 3.--Sample Breakdown. Magazine Time Period Illustrations Total Used Godey's 1840 to 1861 95 92 1862 to 1869 40 40 Peterson's 1842 to 1861 81 67 1862 to 1872 55 52 1873 to 1875 l9 l7 Bazaar 1870 to 1872 10 10 1873 to 1893 90 90 1894 to 1904 55 55 1905 to 1940 175 174 Vogue 1895 to 1904 50 50 1905 to 1940 174 169 4O facial shape--oval, round, or angular hairstyle--short, medium, long,cn1 tOp of head, or on tOp of head with curls hanging free The five body proportions were measured in head- lengths. Table 4 shows the range of values found for each body proportion by time period. All the types of shoulder lepes and facial shapes appeared in all of the time periods. Short hair was not shown from 1840 to 1904, medium hair from 1873 to 1893, long hair from 1894 to 1904 and hair on tOp of the head with curls hanging free from 1894 to 1904 and from 1926 to 1940. The data for the proportions of the Greek figure were a constant which were compared with the mean body pr0portions for each period. The measurements were as follows: Variable Head Lengths total height 7.5 torso length 1.33 waist to floor 5.0 shoulder width 1.5 waist width 1.0 Costume Period: Difference in the Fashion Ideal of Beauty Between Magazines Differences in the fashion ideal of beauty between magazines for the costume periods defined by this study 41 Table 4.--Range of Values Found for Body PrOportions by Costume and Transition Periods. Variable Maximum Minimum Mean SD 1840 to 1861 Total height 12.00 6.50 8.85 .980 Torso length 2.66 1.33 1.87 .294 Waist to floor 8.00 3.66 5.79 .767 Shoulder width 2.50 1.33 1.66 .266 Waist width 1.33 .50 .75 .192 1862 to 1872 Total height 11.50 7.00 8.83 .709 Torso length 2.00 1.00 1.53 .184 Waist to floor 8.50 4.50 6.20 .624 Shoulder width 2.00 1.33 1.53 .162 Waist width 1.66 .50 .78 .271 1873 to 1893 Total height 11.00 7.33 8.68 .679 Torso length 2.00 1.33 1.65 .211 Waist to floor 7.33 4.66 5.91 .582 Shoulder width 2.00 1.00 1.45 .162 Waist width 1.33 .50 .76 .175 Table 4.--Continued. Variable Maximum Minimum Mean SD 1894 to 1904 Total height 11.00 7.00 9.02 .910 Torso length 2.66 1.33 1.99 .308 Waist to floor 7.00 4.33 5.78 .648 Shoulder width 4.00 1.33 1.81 .542 Waist width 1.33 .50 .71 .155 1905 to 1940 Total height 11.66 7.00 8.89 .890 Torso length 3.33 1.33 1.84 .434 Waist to floor 9.00 3.33 5.86 .770 Shoulder width 2.33 1.00 1.54 .224 Waist width 2.00 .66 1.03 .293 43 (see p. 27) were tested as follows: (1) Godey's Lady's Book and Peterson's Magazine, 1842 to 1861; (2) Peterson's Magazine and Harper's Bazaar, 1873 to 1875; and (3) Harper's Bazaar and Egggg, 1905 to 1940. Since there were two types of data which required two types of statistical analyses (see p. 36) multivariate analysis of variance and chi square contingency tables were computed. The multivariate analysis of variance was used to test the difference between the magazines for the means of total heights, torso lengths, waist to floor lengths, shoulder widths, and waist widths. The chi square test was used to determine the differences between magazines for the frequencies of types of shoulder slopes, facial shapes, and hairstyles. Table 5.--Differences Between Magazines, Costume Periods, Multivariate Analyses. Magazine and Time Period F-Ratio Interpretation Godey's, Peterson's 1.135 no difference 1842-1861 Peterson's, Bazaar 1.839 no difference 1873-1875 Bazaar, Vo ue 3.845* difference 1 05-1940 *p<.05 level of significance. 44 For the body prOportions no differences between the magazines were seen from 1840 to 1861, Godey's and Peterson's and from 1873 to 1875, Peterson's and Bazaar (see Table 3). A difference was found between Egggg and Bazaar from 1905 to 1940. The mean body proportions for Bazaar and Vgggg were different for every body proportion; no one prOportion lead to the difference found between the magazines. Mott stated that the policy of 22922 when it was first established required the magazine to use a higher grade of fashion illustrations than other magazines.1 It is quite possible that in "higher grade" fashion illustra- tions there were different ideal fashion figures which might explain the difference in the data from Vgggg and Bazaar from 1905 to 1940. Bazaar did not change into a magazine for high society until after it was purchased by the Hearst Publishing Company in 1913.2 The results of the chi square tests were different from the multivariate analyses used for the body proportions for each costume period (see Table 6). No difference was found in the facial shape between Godey's and Peterson's for 1842 to 1861 but there was a difference between the magazines for the shoulder lepe and hairstyle. The 1Frank Luther Mott, A History of American Magazines, Vol. IV (Cambridge: The Belknap Press of Harvard University Press, 1957), p. 758. 2Mott, A History of American Magazines, Vol. III (1938), p0 3900 45 Table 6.--Differences Between Magazines, Costume Periods, Chi Square. Variable x Interpretation Godey's and Peterson's, 1840 to 1861 Shoulder slope 12.839*** difference Facial shape 7.363 no difference Hairstyle 8.663* difference Peterson's and Bazaar, 1873 to 1875 Shoulder lepe 4.619 Facial shape 21.973*** Hairstyle 4.788 Bazaar and Vogue, 1905 to 1940 Shoulder lepe 3.704 Facial shape 1.488 Hairstyle l7.507*** no difference difference no difference no difference no difference difference *p<.05 level of significance. ***p<.001 level of significance. 46 findings for Peterson's and Bazaar from 1873 to 1875 were just the Opposite. There was a difference in the facial shape between the magazines but no difference between the shoulder slopes or hairstyles. There was nordifference in shoulder lepe or facial shape between Bazaar and Vgggg from 1905 to 1940 but there was a difference in the hair- styles. The lack of consistent findings may mean that there was variability between fashion illustrators in their manner of portraying shoulder slope, facial shape and hairstyle. It was interesting to note that some similarity between Egggg and Bazaar from 1905 to 1940 existed. It may be that during that costume period there was a strong consensus about the ideal type of shoulder slope and facial shape across social class levels which would explain the lack of difference between the magazines. Hypothesis I stated that there would be no differ- ence between fashion magazines in the fashion ideal of beauty during costume periods. For the first costume period, 1842 to 1861, no difference was found between Godey's and Peterson's for body proportions and facial shapes. A difference was found between the magazines for shoulder lepes and hairstyles. The second costume period, 1873 to 1875, showed no difference between Peterson's and Bazaar for body prOpor- tions, shoulder lepes and hairstyles. There was a difference between the magazines for facial shapes. 47 From 1905 to 1940 there were no differences between Egggg and Bazaar for shoulder slope and facial shape but differences were found for body proportions and hairstyles. The differences between magazines in the major characteristics of the fashion ideal of beauty, i.e., the body prOportions, when compared with the differences between magazines in the minor characteristics of the fashion of beauty, i.e., frequencies of types of shoulder slopes, facial shapes and hairstyles were not always the same. There was a difference between Vgggg and Bazaar, 1905 to 1940, for the major characteristics but for two of the minor characteristics, shoulder slope and facial shape there was no difference. The reverse was also true, when no difference in the major characteristics existed between magazines there were differences in some of the minor characteristics between magazines. Costume Transition Period: Difference in the Fashion Ideal of Beauty Between Magazines Differences in the fashion ideal of beauty between magazines for the costume transition periods as defined by this study (see p. 27) were tested as follows: (1) Godey's Lady's Book and Peterson's Magazine, 1862 to 1869; (2) Peterson's Magazine and Harper's Bazaar, 1870 to 1872; and (3) Harper's Bazaar and Vogue, 1894 to 1904. Again, two types of data were collected and required two types of statistical analyses--multivariate analysis of variance for 48 the body prOportions and chi square for the types of shoulder slopes facial shapes and hairstyles. During the costume transition periods no differences were found between Godey's and Peterson's, 1862 to 1869 and Peterson's and Bazaar, 1870 to 1872, but there was a difference between Vgggg and Bazaar, 1894 to 1904 (see Table 7). The mean shoulder width for Bazaar was 1.63 while the mean shoulder width for V9ggg_was 2.00. The mean waist widths were also different; for Bazaar the mean waist width was .73, and for Vpggg it was .70. The other mean body proportions were not different between the magazines. The differences may have been due to the difference in policy between the two magazines (see pp. 39-40). Table 7.--Differences Between Magazines, Transition Periods, Multivariate Analyses. Magazine and Time Period F-Ratio Interpretation Godey's, Peterson's 1.986 no difference 1862 to 1869 Peterson's, Bazaar .658 no difference 1870 to 1872 Bazaar, Vo ue 2.831* difference 1 9 to 1904 *p<.05 level of significance. 49 The chi square tests show that differences existed in the portrayal of shoulder slope, facial shape and hair- style although this again varied over time and between magazines (see Table 8). Between Godey's and Peterson's, 1862 to 1869, there was no difference in the shoulder slope but there was a difference in the facial shape and hair- style. No difference was found between the shoulder slope and hairstyle for Peterson's and Bazaar, 1870 to 1872. Bazaar and Egggg, 1894 to 1904, showed no difference between shoulder slope, facial shape and hairstyle. From 1894 to 1904 the ideal types of shoulder slope, facial shape and hairstyle may have been so much the norm that little difference occurred between any fashion illustrations. The turn of the century was the time period in which Charles Dana Gibson's illustrations were widely published and the effect of his "Gibson girl" could have created a well accepted fashion ideal of beauty. Hypothesis II stated that there would be no differ- ences between fashion magazines in the fashion ideal of beauty during any costume transition period. During the first costume transition period, 1862 to 1872, no difference was found between Godey's and Peterson's for body prOportions and shoulder slope and between Peterson's and Bazaar for body proportions, shoulder lepe and hairstyle. A difference was found between Godey's and Peterson's for facial shape and hairstyle and between Peterson's and Bazaar for facial shapes. 50 Table 8.--Differences Between Magazines, Transition Periods, Chi Square. Variable X Interpretation Godey's and Peterson's, 1862 to 1869 Shoulder lepe 1.222 no difference Facial shape l7.224*** difference Hairstyle ll.ll6** difference Peterson's and Bazaar, 1870 to 1872 Shoulder slope .816 no difference Facial shape 11.359*** difference Hairstyle .490 no difference Bazaar and Vogue, 1894 to 1904 Shoulder .044 no difference Facial shape 1.681 no difference Hairstyle 1.111 no difference **p<.02 level of significance. ***p<.01 level of significance. 51 For the second costume transition period, 1894 to 1904, no difference was found between Bazaar and Veggg for shoulder slope, facial shape and hairstyle. A difference was found between the magazines for body proportions. The major characteristics of the fashion ideal of beauty showed no differences between magazines from 1842 to 1875. The three minor characteristics were not consistent over time or between magazines. There was no difference in the frequencies of shoulder lepes between magazines from 1862 to 1940. Frequencies of facial shapes were not different between magazines from 1841 to 1862 and from 1894 to 1940 while frequencies of hairstyles were not different between magazines from 1870 to 1940. Costume Peripd: Change in the Fashion Ideal of Beauty Change in the fashion ideal of beauty for the costume periods as defined by this study (see p. 27) was first tested as follows: (1) 1840 to 1861, (2) 1873 to 1893, and (3) 1905 to 1940. Contrary to the stability which was hypothesized, change occurred within all the costume periods (see Table 9). The years of each period were divided and tested with a multivariate analysis to determine if the time of the change could be isolated. The second division of years were as follows: 52 Table 7.--Change in the Fashion Ideal of Beauty Within Costume Periods, Multivariate Analyses. Time Period F-ratio Interpretation 1840 to 1861 2.392* change 1840 to 1846 3.751** change 1847 to 1851 1.588 no change 1852 to 1855 no data 1856 to 1861 6.875* change 1873 to 1893 7.472*** change 1873 to 1879 2.008* change 1880 to 1886 2.109* change 1887 to 1893 no data 1905 to 1940 18.758*** change Bazaar 1905 to 1911 3.454** change 1912 to 1918 3.851** change 1919 to 1925 4.949** change 1926 to 1932 .285 no change 1933 to 1940 2.422 no change 1905 to 1940 2.524** change Vogue 1905 to 1911 4.187*** change 1912 to 1918 .987 no change 1919 to 1925 12.855*** change 1926 to 1932 1.858 no change 1933 to 1940 2.249* change *p<.05 level of significance. **p<.01 level of significance. ***p<.001 level of significance. 53 l. 1840 to 1862 a. 1840 to 1846 b. 1847 to 1851 c. no data for 1852 to 1855 d. 1856 to 1861 2. 1873 to 1893 a. 1873 to 1879 b. 1880 to 1886 c. no data for 1887 to 1893 3. 1904 to 1940 a. 1904 to 1911 b. 1912 to 1918 c. 1919 to 1925 d. 1926 to 1932 e. 1933 to 1940 The chi square was tested as follows: 1. 1840 to 1861 2. 1873 to 1893 3. 1904 to 1925 4. 1926 to 1940 Both multivariate analyses and chi square tests were per- formed. During the first costume period, 1840 to 1861, change occurred from 1840 to 1846 but no change from 1847 to 1851; all other groups of years within that time period showed change. The fact of no change in the middle of the period partially supported the hypothesis in this case. 54 However, the pattern did not hold true for either of the following periods, 1873 to 1893 or 1905 to 1940. There was evidence of changes occurring throughout the second costume period, 1873 to 1893. The third period, 1905 to 1940 was tested for change in the fashion ideal of beauty separately for each magazine since a difference in fashion ideals was found between the magazines. For Bazaar no change was found from 1926 to 1940 in the body proportions of the fashion ideal of beauty, although change was found from 1905 to 1925. No change was found in the body prOportions of the fashion ideal of beauty as portrayed in Vgggg in the period of years from 1912 to 1918 and from 1926 to 1932. Change was found from 1905 to 1925 for Bazaar and from 1905 to 1911, 1919 to 1925, and 1933 to 1940 for Vgggg. No change was found for the shoulder slope from 1840 to 1861 and from 1925 to 1940; there was change in the shoulder lepe from 1873 to 1893 and from 1905 to 1925 (see Table 10). The facial shape showed change from 1840 to 1861 and 1873 to 1893 and did not show change from 1905 to 1940. The hairstyle showed change in every period. Looking at the results one can notice that there is a difference between change in the body prOportions and change in the frequency of shoulder lepes, facial shapes and hairstyles. Some of this difference may be due to the differences in the two types of data and statistical analyses. It is also possible that the body proportions of the fashion ideal of beauty may change at different 55 Table 10.--Change in the Fashion Ideal of Beauty Within Costume Periods, Chi Square. Time Period/ Variable X2 Interpretation 1840 to 1861 Shoulder slope 11.144 no change Facial shape 77.036*** change Hairstyle 68.059*** change 1873 to 1893 Shoulder lepe 28.968*** no change Facial shape 16.492* change Hairstyle 29.425*** change 1905 to 1925 Shoulder slope 21.897* change Facial shape 15.092 no change Hairstyle 24l.014*** change 1926 to 1940 Shoulder slope 6.947 no change Facial shape 2.224 no change Hairstyle l9.834*** change *p<.01 level of significance. **p<.02 level of significance. ***p<.001 level of significance. 56 times from changes in shoulder lepes, facial shapes and hairstyles. Young randomly selected illustrations from each year of her study and determined the most frequent garment styles for each year. This study randomly selected illustrations by Young's costume silhouette periods, and mean body prOportions were calculated for each of the costume periods. This could account for the results being different from what was hypothesized since this study did not duplicate Young's methods. Hypothesis III stated that there would be no change in the fashion ideal of beauty during costume periods as defined by Young. For the first period, 1840 to 1861, there was no change in body proportions from 1847 to 1851; change was found from 1840 to 1846 and from 1852 to 1861. No change in the frequency of types of shoulder SIOpe was found for this period although change was seen in the facial shape and hairstyle. During the second period, 1873 to 1893, change was found in the body prOportions, facial shape and hairstyle. There was no change in the frequency of types of shoulder lepes. The third period, 1905 to 1940, showed no change in body prOportions for Bazaar from 1926 to 1940 and for Egggg from 1912 to 1918 and 1933 to 1940 and for frequency of facial shapes. Change was found for shoulder slopes and hairstyles from 1905 to 1925. No change in the frequency of shoulder slopes was evident from 1926 to 1940. The first two periods, 1840 to 1861 and 1873 to 1893, showed no change in the shoulder slope and a change 57 in facial shape and hairstyle. No other pattern was evident. There was no consistency in change found for the major characteristics compared with change in the minor characteristics. Costume Transition Period: Change in the FashIOn Ideal of Beauty Change in the fashion ideal of beauty for the costume transition periods as defined by this study (see p. 27) was tested by breaking the periods into five groups of overlapping years. The first transition period, 1862 to 1872, was divided as follows: (1) 1862 to 1864, (2) 1864 to 1866, (3) 1866 to 1868, (4) 1868 to 1870, and (5) 1870 to 1872; and the second transition period: (1) 1894 to 1896, (2) 1896 to 1898, (3) 1898 to 1900, (4) 1900 to 1902, and (5) 1902 to 1904. Again both a multivariate analyses and chi square tests were performed because of the two types of data. Change in the mean body proportions of the fashion ideal of beauty was hypothesized during the transition periods but was not found to be true from 1862 to 1872. From 1894 to 1904 there was one period of change revealed from 1898 to 1900 (see Table 11). Only shoulder slope and facial shape from 1862 to 1872 showed no change. Hairstyle for both transition periods and shoulder slope and facial shape from 1894 to 1904 changed (see Table 12). 58 Table ll.--Change in the Fashion Ideal Of Beauty Within Transition Periods, Multivariate Analyses. Time Period F-ratio Interpretation 1862 to 1872 1862 to 1864 .790 no change 1864 to 1866 1.263 no change 1866 to 1868 1.441 no change 1868 to 1870 1.423 no change 1870 to 1872 .810 no change 1894 to 1904 1894 to 1896 .769 no change 1896 to 1898 1.216 no change 1898 to 1900 2.521* change 1900 to 1902 1.075 no change 1902 to 1904 .978 . no change *p<.05 level of significance. Comparing these results with the results from the costume periods (see Tables 9, 10) one can note that there were no changes in frequencies of types of shoulder slope from 1840 to 1893, and there were changes in the frequencies of types of hairstyles during every period, 1840 to 1940. The difference in the results between the two types of data could be due to the fact that each process is measuring different variables. The fashiOn ideal of beauty, based on the literature, appeared to reflect change in body proportions while the frequency counts of the occurrence of like characteristics may reflect popularity of shape. 59 Table 12.--Change in the Fashion Ideal of Beauty Within Transition Periods, Chi Square. Time Period/ Variable X2 Interpretation 1862 to 1872 Shoulder lepe 3.124 no change Facial shape 3.279 no change Hairstyle 17.418* change 1894 to 1904 Shoulder slope 18.324** change Facial shape 26.139*** change Hairstyle 105.000*** change *p<.05 level of significance. **p<.02 level of significance. ***p<.001 level of significance. It may be that during each costume period and transition period different parts of the fashion ideal of beauty are of importance and may be emphasized or de- emphasized. These areas of emphasis could change even within one period depending on small costume style changes. Thus by looking at the fashion ideal of beauty over long periods of time it may not be possible to discover the change by using a limited set of measures. It is also possible that the methods used in this study were not compatible with the costume silhouette periods determined by Young due to the fact that she used different research 60 methods to obtain her results. It might be necessary to include measures that would relate to skirt silhouettes, such as hip width. Briefly, Hypothesis IV stated that change would occur during transition periods in the body prOportions of the fashion ideal of beauty and in the frequencies of types of shoulder lepes, facial shapes, and hairstyles. During the first period, 1862 to 1872, there was no change in the mean body proportions or frequencies of shoulder slope and facial shape. There was change for the hairstyles. The second period, 1894 to 1904, showed change in mean body prOportions from 1898 to 1900 and in frequencies of shoulder lepe, facial shape and hairstyle from 1894 to 1904. No change in body prOporations was found from 1894 to 1898 and from 1900 to 1904. Difference Between the Classic Greek Figure and the FashIOn Ideal of Beauty A univariate analysis of variance was used to deter- mine if there was a difference between the classic Greek figure and the fashion ideal of beauty during any costume period or costume transition period as defined by this study (see p. 38). Table 13 shows the results of the tests. With the exception of the mean shoulder width from 1862 to 1872 and the mean waist width from 1905 to 1940 there were no similarities between the Greek standard of beauty and the fashion ideal of beauty. Since the two similarities were out of a possible twenty-five, there does 61 Table l3.--Differences Between the Fashion Ideal of Beauty and the Greek Standard of Beauty, One Way Analysis of Variance. Variable F-ratio Interpretation 1840 to 1861 Total height 302.024*** difference Torso length 543.190*** difference Waist to floor 170.537*** difference Shoulder width 61.409*** difference Waist width 273.561*** difference 1862 to 1872 Total height 351.074*** difference Torso length 118.262*** difference Waist to floor 368.102*** difference Shoulder width 2.727 no difference Waist width 63.251*** difference 1873 to 1893 Total height 321.7l9*** difference Torso length 251.191*** difference Waist to floor 260.811*** difference Shoulder width 10.608*** difference Waist width 205.825*** difference Table l3.--Continued. 62 Variable F-ratio Interpretation 1894 to 1904 Total height 29l.694*** difference Torso length 486.558*** difference Waist to floor 153.402*** difference Shoulder width 33.886*** difference Waist width 361.908*** difference 1905 to 1940 Total height 846.497*** difference Torso length 471.101*** difference Waist to floor 434.835*** difference Shoulder width 12.137*** difference Waist width 3.193 no difference ***p<.001 level of significance. 63 not seem to be a relationship between the Greek standard of beauty and the fashion ideal of beauty. The hypothesis stated that there would be a differ- ence between the fashion ideal Of beauty and the Greek standard of beauty during any costume period or any transition period. The shoulder width from 1862 to 1872 and the waist width from 1905 to 1940 showed no difference; all other body prOportions and all other periods showed a difference. Summary For the first costume period, 1840 to 1861, no difference was found between Godey's and Peterson's for mean body prOportions and frequencies of hairstyles. No change was found in body proportions from 1847 to 1851 and in frequencies of the shoulder lepe for the entire period. There was a difference between the mean body prOportions for the period and the Greek standard of beauty. During the first costume transition period, 1862 to 1872, there was no difference between Godey's and Peterson's, 1862 to 1869, and Peterson's and Bazaar, 1870 to 1872, for mean body proportions and frequencies of shoulder lepes. There was no difference between Peterson's and Bazaar for frequencies of hairstyle. No change was seen in body prOportions for the entire period, however, there was a change in the frequencies of hairstyle. With the exception of shoulder width there was a difference between the mean 64 body prOportions for the period and the Greek standard of beauty. The second costume period, 1873 to 1893, showed no difference in body proportions or frequencies of shoulder slope and hairstyle between Peterson's and Bazaar, 1873 to 1875. Change was seen in the body prOportions for the entire period, however, no change was evident in the frequencies of shoulder slope. There was a difference between the mean body prOportions for the entire period and the Greek standard of beauty. For the second transition period, 1894 to 1904, there was no difference between Vgggg and Bazaar for fre- quencies of shoulder lepes, facial shapes and hairstyles. There was change in the body prOportions from 1898 to 1900 and change in the frequencies of shoulder slope, facial shape and hairstyles for the entire period. A difference was found between the mean body proportions for the entire period and the Greek standard of beauty. During the third costume period, 1905 to 1940, no difference between Veggg and Bazaar was found for fre- quencies of shoulder slope and facial shape. No change was indicated in the body proportions measured in Egggg from 1926 to 1940 and in Bazaar from 1912 to 1918 and from 1926 to 1932. There was also no change in frequencies of facial shape from 1905 to 1940 and shoulder slope from 1926 to 1940. With the exception of the waist width there was a 65 difference between the mean body prOportions for the period and the Greek standard of beauty. Beginning in 1842 through 1875 there was no differ- ence in the major characteristics between the magazines. From 1862 to 1940 there was no difference in the frequency of shoulder slopes between magazines, and from 1894 to 1940 there was no difference in the frequency of facial shapes between magazines. There was no difference in the frequency of hairstyles between magazines from 1870 to 1904. From 1840 to 1893 there was no change in the frequencies of shoulder lepes, and from 1840 to 1940 there was change in the frequency of hairstyles. CHAPTER VI SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Summary A comparison of fashion magazines disclosed that there is a visual difference in the portrayal of fashion illustrations over the years. From a review of the literature it became apparent that a standard of beauty with body prOportions modeled after classic Greek sculpture existed as well as an ideal of beauty illustrated in fashion magazines which varied from time to time. There seemed to be a need for research to identify and define different ideals of beauty and to relate them to costume periods. A decision was made to use Young's costume silhou- ette periods as the basis for the expectation of change in the fashion ideal of beauty.1 Ten year transition periods were used for the period of years during the transition of costume silhouettes. Fashion magazines were used as a data source and Godey's, Peterson's, Bazaar, and Vogue were selected for the study which included the years 1840 to 1940. lYoung, Recurring Cycles of Fashion, p. 20. 66 67 Magazine issues were randomly selected from the costume periods and a random sample was taken for each year of the transition periods. The first five full front illustrations were selected and measured by headlength for total height, torso length, waist to floor length, shoulder width, and waist width. Shoulder lepe, facial shape and hairstyle types were also noted. There were five main hypotheses: (1) there would be no difference in the fashion ideal of beauty between the two fashion magazines used for each costume period, (2) there would be no difference in the fashion ideal of beauty between the two fashion magazines used for each silhouette period, (3) there would be no change in the fashion ideal of beauty during costume periods, (4) there would be change in the fashion ideal of beauty during the transition periods, and (5) there would be a difference between the fashion ideal of beauty for any one period and the Greek standard of beauty. Univariate and multivariate analyses of variance were used to test for the differences between the magazines and the change during the periods for the body prOportions. Chi square contingency tables were used to determine the frequency of the types of shoulder slope, facial shape, and hairstyle. None of the major hypotheses were totally unsupported since each hypotheses was supported in at least one costume period and/or transition period. 68 Conclusions There was no difference found between Godey's and Peterson's for 1842 to 1869 and Peterson's and Bazaar for 1870 to 1875. This shows some evidence of like portrayal of model styles between fashion magazines. One cannot con- clude that this is always true since a difference was found between Egggg_and Bazaar from 1894 to 1940. Further research needs to be done before a conclusive statement can be formulated. There seems to be some evidence of a relationship between the fashion ideal of beauty and the shoulder slope, facial shape, and hairstyle due to the fact that these characteristics showed evidence of change in frequency of appearance during costume transition periods and stability during costume periods. It is possible, since the body prOportions of the fashion ideal of beauty did not change as hypothesized, that changes in the fashion ideal of beauty might not be related to or correspond with changes in skirt silhouettes as determined by Young. Another explanation for the results might be due to the fact that although the study made use of Young's costume silhouette periods the procedure used was not the same as Young's. Young randomly selected illustrations from each year of her study, determined frequencies of garment characteristics for each year then made a composite sketch combining these characteristics, and determined the skirt silhouette periods by viewing the composite 69 illustrations placed in chronological order. This study used a random sample divided into silhouette periods according to Young's findings. Means of the body prOpor- tions were determined for each period. It might be necessary to duplicate the procedure followed by Young; to make composite sketches of the most frequently occurring body features of the fashion ideal of beauty and then visually determine when change occurs. Greek figure prOportions did not appear to be the guide for depicting the beautiful in fashion illustrations, and, based on the results of this study any relationships found between the fashion ideal of beauty and the Greek standard of beauty are probably due to chance (see Implications section below). The introduction of new printing techniques in the late nineteenth and early twentieth century was not investigated in this study. These new techniques gave illustrators more freedom in terms of possible drawing styles and undoubtedly may have affected styles of fashion figures. Use of the new techniques may have been reSponsible for some of the unpredicted findings in this study and should be considered in future research. Vari- ations in styles of illustrators must also be considered. It is a historical fact that certain pOpular illustrators such as Gibson or Erté had their own distinctive styles which other illustrators copied extensively. 70 One interesting outcome was that there appeared to be little difference in the manner of illustrating fashion figures between European and American fashion plates from 1842 to 1869 since there was no difference seen between the illustrations in Peterson's Magazine and Godey's Lady's Book. Godey's published American fashion plates while Peterson's published European fashion plates. It is possible that American and European illustrators had similar training. It must also be noted that this study did not attempt to investigate color or the possible influence color may have had on the portrayal of the fashion ideal of beauty. Color no doubt played a part in the reader's perception of the fashion ideal of beauty as it would illustrate the type of complexion, color of hair, etc. However, it was not possible to study color because photo- COpies of illustrations were used in determining measure- ments. Also fashion plates were rarely colored exactly the same from COpy to copy; it would have been impossible to study all the aspects of colored illustrations. Implications Possibly the most significant finding of this study was the discovery that with the exception of shoulder width from 1862 to 1872 and waist width from 1905 to 1940 there was a difference between the mean body prOportions of the 71 fashion ideal of beauty and the Greek standard of beauty for every period from 1840 to 1940. As mentioned in the review of literature, authors of clothing selection books have been using the Greek standard of beauty as an ideal, and students are taught to manipulate clothing design lines to achieve the illusion of the standard. This study has shown that this standard does not, for the years 1840 to 1940, coincide with the several fashion ideals of beauty. Perhaps, students of the aesthetics of dress would find it more satisfying to try to achieve the currently fashionable figure through the use of illusionary lines and forms. What appears to be needed is to recognize the existence of the two ideals of beauty. Students should learn to be sensitive to the beauty of the proportions of the Greek standard figure. It seems that not only is there a need for students of art and clothing selection to be taught the proportions of the Greek standard, but they should also be made aware of the fashion ideal of beauty as developed and illustrated by fashion artists and its role in the clothing world. Recommendations re study needs to be done before definite con- clusion can be made concerning the fashion ideal of beauty. The following are some of the areas that need investigation: 72 Smaller periods of time need to be investigated by randomly sampling from each year of the study to find the relationship between change in the fashion ideal of beauty and various garment style changes. The measuring techniques used in this study need to be modified. Possibly other body characteristics than shoulder lepe, facial shape and hairstyle need to be included. Instead of measuring body prOportions the prOportions could be classified by types of figure, i.e., hourglass, etc., and then counted for frequency of occurrence. Differences in illustrator's styles and differences in the fashion ideal of beauty should be studied. Different illustrators have different styles which may create a distinctly personal ideal of beauty which can become fashionable. A comparison of each well known illustrator's works should be made to determine what differences do exist between illustrators. For the magazines published in the twentieth century differences in the illustrative techniques have been available. These may or may not have affected the portrayal of the fashion ideal of beauty. A study is needed to determine types of techniques used and differences in the fashion ideal Of beauty by illustrative techniques. 73 In this study Young's skirt silhouette periods were used without duplicating any of her research methods. The results showed that the fashion ideal of beauty did not coincide with her costume periods which were based on skirt silhouettes. A study is needed which would duplicate Young's research and would include other garment features as well. It is necessary to discover if Young's periods are accurate only for skirt widths. Research is also needed that would study the rate of change in clothing styles and in the fashion ideal of beauty. It is quite possible that various garment features and parts of the fashion ideal of beauty change at similar rates but are not related to the more evolutionary change in skirt silhou- ettes. These differences need to be known for a greater awareness of the fashion process. Another study that would be of benefit to under- standing the process of style change would be an in depth study of one costume silhouette period. More data could be obtained for each year which would make it possible to analyze change from year to year. It is also recommended that whenever possible a random sample which selects data from each year of the study be used for the above studies. This 74 would allow differences from year to year to be found more easily. BIBLIOGRAPHY BIBLIOGRAPHY Books // L/Ahspach, Karlyne. The Why of Fashion. Ames, Iowa: The Iowa State University Press, 1967. Bigelow, Marybelle S. Fashion in History; Apparel in the Western World. Minneapolis: Burgess Publishing Company, 1970. Braun-Ronsdorf, Margarete. Mirror of Fashion. Translated by Oliver Coburn. New York: McGraw Hill Book Company, 1964. Buttrick, Helen Goodrich. Principles of Clothing Selection. \/ New York. The Macmillan Company, 1925. Chambers, Helen G., and Moulton, Verna. Clothing Selection. Philadelphia: J. B. Lippincott Company, 1969. Cunnington, C. Willett. Why Women Wear Clothes. London: Faber and Faber, Ltd., 1941. Flfigel, J. C. The PsycholOgy of Clothes. London: Hogarth Press LTD, 1930. Hartley, Jonathan Scott. Anatomy in Art. New York: Styles and Cash, 1891. Hillhouse, Marion S. Dress Selection and Design. New York: The Macmillan Company, 1963. Jackson, Arthur Gladstone. The Right Way to Human Figure \" Draw and Anatomy. New York: Emerson Books, Inc., 1952. Kohler, Carl. A History of Costume. New York: Dover Publications, Inc., 1963. .4". ~ Laver, James. Clothes. New York: Horizon Press, 1953. 75 76 Morton, Grace Margaret. The Arts of Costume and Personal Appearance. Revisedfiby Mary E. Guthrie, VilIetta Leite, and June Ericson. New York: John Wiley and Sons, Inc., 1964. Mott, Frank Luther. A History of American Magazines. Vol. I through IV. Cambr1dge: The Belknap Press of Harvard University Press, 1938- 1968. Nystrom, Paul H. Economics of Fashion. New York: The Ronald Press Co., 1928. Payne, Blanche. History_of Costume. New York: Harper and Row, Publishers, Inc., 1965. fl Roach, Mary Ellen, and Eicher, Joanne B. The Visible Self. \' Englewood Cliffs, N.J. : Prentice—Ha11,Inc., 1973. Thompson, Arthur. Handbook of Anatomy for Art Students. Oxford: The Clarendon Press, n.d. Young, Agatha Brooks. Recurring Cycles of Fashion. New York. C00per Square Publishers, Inc., 1966. Periodicals Godey's Lady's Book. Philadelphia: 1840 to 1869. Harper's Bazaar. New York: 1870 to 1940. Peterson's Magazine. Philadelphia: 1862 to 1875. Vogue. New York: 1895 to 1940. Kroeber, A. L. "On the Principle of Order in Civilization as Exemplified by Changes of Fashion." American Anthropologist, XXI (July, 1919), 235-63. Pitz, Henry C. "Millions of Pictures." American Artist (November, December, 1961), 35-39, 76-78, and 52-57, 85. Richardson, Jane, and Kroeber, A. L. "Three Centuries of Women's Dress Fashions, A Quantitative Analysis." Anthropological Records, Vol. 5, No. 2 (1940), 111-53. 77 Unpublished Materials Hackler, Ruth Nadine. "Fashion Cycles in Style, Fabric, and Design of Women's Skirts in Four Fashion Magazines from 1925-1961." Unpublished M.S. thesis, Oklahoma State University, 1962. Joseph, Marjory Lockwood. "Changes in Women's Daytime Dress as Related to Other Selected Cultural Factors During the First and the Third Decades of the Twentieth Century." Unpublished Ph.D. dissertation, Pennsylvania State University, 1962. Kelley, Carole Louise. "An Analysis of Two Theories of Fashion Change During the Years 1927-1934." Unpublished M.S. thesis, Cornell University, 1970. Malory, Mary Ann. "The Relationship Between Fluctuations in Hemlines and Stock Market Averages From 1921 to 1971." Unpublished M.A. thesis, University of Tennessee, 1971. Tortora, Phyllis G. "The Evolution of the American Fashion Magazine as Exemplified in Selected Fashion Journals 1830-1969." Unpublished Ph.D. dissertation, New York University, 1973. General References Avedon, Richard. "Harper's Bazaar." Graphis (1968), 258-69. Basralian, Karen Marie. "The American Woman's Day Dress, 1890 Through 1900, as Reflected in the American Fashion Magazines." Unpublished M.S. thesis, University of Maryland, 1969. Beaton, Cecil. Fashion. London: Victoria and Albert Museum, 1971. Bell, Quentin. On Human Finery. London: The Hogarth Press, 1948. Chambers, Bernice. Fashion Fundamentals. New York: Prentice-Hall, Inc. 1947. Dilley, Romilda. Fundamental Fashion Drawing. New York: Sterling Publishing Co., 1967. Finley, Ruth B. The Lady of Godeyfs Sarah Joseph Hale. Philadelphia: J. B. Lippincott Co., 1931. 78 Hall, Carrie A. From Hoopskirts to Nudity. Caldwell, Idaho: The Caxton Printers, Ltd., 1938. Hawes, Elizabeth. Fashion Is Spinach. New York: Random House, 1938. Hornback, Betty C. "The Influences Upon and Identifying Criteria of Women's Fashion in the United States from 1939—1948." Unpublished M.S. thesis, Indiana University, 1967. Jarnow, Jeannette A., and Judelle, Beatrice. Inside the Fashion Business. New York: John Wiley afia Sons, Inc., 1965. Lewis, Benjamin M. An Introduction to American Magazines 1800-1810. Ann Arbor, Mich.: University of Michigan, Department of Library Science Studies, 1961. Lumbra, Elaine L. "Every-Day Women's Dresses Worn in the United States During the Period 1870-1900." Un- published M.S. thesis, Indiana University, 1966. Matthews, Linda Mae. "Changes in Men's Business Suit as Reflected in Fashion Advertisements 1901 Through 1950." Unpublished M.S. thesis, University of Maryland, 1969. Merrifield, Mrs. Dress as a Fine Art. Boston: John P. Jewett and Co., 1854. Nevinson, John L. Origin and Earlnyistory of the Fashion Plate. U.S. National Museum Bulletin 250, Paper 60. Washington, D.C.: Smithsonian Press, 1967, pp. 65-92. Stone, Bernard, and Eckstein, Arthur. PreparinggArt for Printing. New York: Reinhold PubliShing Co., 1965. Stroup, Ann Helen. "An Investigation of the Dress of American Children from 1930 Through 1941 with Emphasis on Factors Influencing Change." Unpub- lished M.S. thesis, Cornell University, 1967. Traphagen, Ethel. Costume Design and Illustration. New 1 York: John Wiley and Sons, Inc., 1918. Waite, Emma Forbes. "Early Artists of Godey's Lady's Book." Antigues (April, 1951), 300-01. 79 Weitenkampf. "Nineteenth Century History in Illustrated Weeklies." Art in America, Vol. 54, No. 4, 28-31. White, Elizabeth Ann. "Silhouette Characteristics of Windle, Women's Daydresses in the United States from 1900-1920." Unpublished M.A. thesis, University of Iowa, 1970. B. C. A. Proportions of the Body. London: Bailliere, Tindall and Cox, 1892. APPENDICES APPENDIX A METHOD OF MEASURING ILLUSTRATIONS APPENDIX A METHOD OF MEASURING ILLUSTRATIONS I. Total Height* A. Determine tOp of the head 1. Find the distance from the center of the eyes to the chin 2. Add the above amount (I.A.1) from the center of the eyes upward E. Find the tip of the toes, if not visible use the edge of the hem 1. Use a straight edge to determine the end of the figure 2. Place a straight edge perpendicular to the figure and move it downward on the figure to the point where it touches only the end of the figure or the edge of the skirt hem 3. Measure from the end of the figure (1.8.2) to the tOp of the head (I.A.2) *Use the left side of the figure unless it is not visible in the illustration. 80 1 unit 1 head { am; I.A.2 II. Torso Length A. Determine the top of the shoulder 1. Find the point where the shoulder line meets the neckline a. extend the shoulder line to the point where it meets the line of the neck b. extend the line of the neck to the point where it meets the shoulder line B. Determine the waistline 1. Use waistline seam for waistline 2. If there is no waistline seam use the waist indentation 82 3. If there is neither of the above use the first design line below the bust area for the waistline C. Measure torso length 1. Draw a line perpendicular to the top of the shoulder (II.A.1.c.) 2. Measure distance between the t0p of the shoulder (II.A.1.c.) and the waist (II.B.1) ILAJ II.C II.C [I.B.2 :EFJ 11.3.3 ) Measure distance between the waistline (II.B.) III. Waist to Floor and the end of the figure (I.B.) along a line which is perpendicular to both I.B. and II.B. 83 III IV. Shoulder Width A. Draw a line perpendicular to the end of the figure line (I.B.) along the outside edge of each shoulder B. Measure horizontally between the two lines IV.B V. I. V. 84 V. Waist Width Determine waistline (II.B.) and measure width of the waist by measuring between the outside lines of the figure at the waistline VI. Shoulder Slope A. Use the shoulder slope line (II.A.1.a.) B. Extend a line perpendicular to the figure from the point where the shoulder line meets the line of the neck C. Measure the angle formed by the shoulder slope line (II.A.1.a.) and the line perpendicular to the figure which extends outward from the neck (VI.B.) \ I.A.1.a 85 VII. Facial Shape A. Compare the face of the illustrations with the drawings of face shapes B. Determine whether the face shape is round, oval or has a broad forehead with a pointed chin A-—oval B--round C--broad forehead with a pointed chin VIII. Hair Style A. Compare the hairstyle of the illustrations with the drawings of hairstyles B. Determine whether the hair is short, medium, long, on tOp of the head, or both on tOp of the head and hanging free 1. Short-~ear length hair or shorter 2. Medium--hair length is between the ears and the chin 3. Long--hair length is below the chin 4. Hair done on top of the head--there is no hair hanging down 5. Hair done on top of head with some hair hanging free 87 oz coflpmupmsHHH wamumuflmm mmmnm Hafiomm mmoam Hmoasonm gumcmq xomz Buoflz umflmz SDUHZ HOUHSOSm hooflm\umamz cumcmq omuoe unmflmm Hmuoe muwo ..oz mwmm ..oz ..Ho> umcwummmz 88 Vggue, Vol. 37, #2, p. 46. I??? ' - W ‘ I 'I’i' : |./“t' ex: 462"" {‘1 1' ._(3J _‘ A THESE SMART NEW MODELS OF CREPE. LINEN AND SERGE ARE DESIGNED FOR THE SOUTHERN SEASON has on b u .10 APPENDIX B SELECTION OF DATA APPENDIX B SELECTION OF DATA Godey's Lady's Book Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume 20, Numbers 1, 2, 4, 5; January, February, April, May, 1840 21, Numbers 1, 2, 3, 4; July, August, September, October, 1840 22, Numbers 3, 4, 6; March, April, June, 1841 24, Numbers 3, 5; March, May, 1842 25, Numbers 1, 3, 4; July, September, October, 1842 26, Numbers 1, 2, 3, 4; January, February, March, April, 1843 27, Numbers 1, 3; June, September, 1843 28, Numbers 2, 3; February, March, 1844 29, Numbers 3, 6; September, December, 1844 30, Numbers 3, 5, 6; March, April, June, 1845 31, Number 1, July, 1845 32, Numbers 3, 4; March, April, 1846 33, Numbers 2, 3, 4, 5, 6; August, September, October, November, December, 1846 34, Numbers 5, 6; May, June, 1847 89 Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume 90 35, Numbers 1, 3, 4, 5; July, September, October, November, 1847 36, Number 1, January, 1848 38, Numbers 1, 2, 3, 4; January, February, March, April, 1849 40, Number 3; March, 1850 41, Numbers 1, 2; July, August, 1850 55, Numbers 4, 6; October, December, 1857 57, Numbers 1, 2; July, August, 1858 65, Numbers 2, 3, 4; August, September, October, 1862 67, Numbers 3, 4, 5, 6; September, October, November, December, 1863 68, Numbers 3, 6; March, June, 1864 70, Number 2; February, 1865 73, Numbers 2, 3, 4; August, September, October, 1866 75, Numbers 5, 6; November, December, 1867 77, Numbers 4, 6; October, December, 1868 79, Numbers 4, 6; October, December, 1869 Peterson's Magazine Volume Volume Volume Volume Volume Volume Volume 1, Numbers 2, Numbers Numbers Numbers , Numbers 1. , Number 2; February, June, 1842 October, December, 1842 5, 6; April, May, June, 1843 6; October, November, December, 1843 4, 5; January, April, May, 1844 Number 5; May, 1845 August, 1845 Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume 91 9, Number 5; May, 1846 14, Numbers 2, 3, 5; August, September, November, 1848 15, Number 4; April, 1849 16, Numbers 2, 3, 4, 5, 6; August, September, October, November, December, 1849 23, Numbers 1, 2, 3, 5, 6; January, February, March, May, June, 1853 25, Number 6; June, 1854 26, Numbers 2, 3; August, September, 1854 35, Number 3; March, 1859 36, Number 1; August, 1859 37, Number 6; June, 1860 38, Number 5; November, 1860 39, Number 6; June, 1861 40, Numbers 1, 2, 3; July, August, September, 1861 42, Numbers 2, 3, 4; August, September, October, 1862 43, Number 3; March, 1863 44, Numbers 2, 6; August, December, 1863 46, Numbers 2, 5, 6; August, November, December, 1864 48, Numbers 2, 3, 4, 5, 6; August, September, October, November, December, 1865 50, Numbers 3, 4; September, October, 1866 52, Numbers 1, 2, 6; July, August, December, 1867 53, Number 6; June, 1868 54, Numbers 1, 2, 4, 6; July, August, October, December, 1868 55, Numbers 3, 4; March, April, 1869 Volume 58, Numbers 2, Volume 60, Numbers 3, 1871 Volume 61, Number 6; June, 92 3; August, September, 1870 5, 6; September, November, December, 1872 Volume 62, Number 2; August, 1872 Volume 63, Numbers 2, VOlume 64, Numbers 1, Volume 66, Numbers 1, Volume 67, Numbers 3, Volume 68, Harper's Bazaar Volume 3, Numbers 5, Number 1; July, 7. 4; February, April, 1873 2; July, August, 1873 2, 3; July, August, September, 1874 4, 6; March, April, June, 1875 1875 9, 31, 32, 35; January, February, July, and August, 1870 Volume 7, Numbers 7, 8. 9, 26, 28, 32, 34, 35; February, July, August, 1874 Volume 8, Numbers 48, Volume 9, Numbers 33, Volume 10, Numbers 4, Volume 12, Numbers 24, Volume 13, Numbers 49, Volume 15, Numbers 29, October, 1882 Volume 16, Numbers 34, December, 1883 Volume 17, Numbers 27, Volume 19, Numbers 50, 49, 51, 52; November, December, 1875 34, 35, 36; August, September, 1876 5; January, February, 1877 25, 27; June, July, 1879 50; December, 1880 30, 31, 33, 42, 43; July, August, 51, 52; August, September, July, 1884 December, 1886 Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume 21, 22, 27, 28, 29, 30, 31, 32, 33, 35, 36, 37, 38, 39, September, October, 40, 41, 42, 43, 45, 46, 47, 48, 51, 54, 93 Numbers 17, 19; April, May, 1888 Numbers 6, 7; February, 1889 Numbers 8, 9; February, 1894 Number 4; January, 1895 Number 34; August, 1896 Number 36; September, 1897 Numbers 32, 33; August, 1898 Numbers 29, 30; July, 1899 Numbers 35, 36; September, 1900 Number 2; June, 1901 Number 5; May, 1902 Number 10; October, 1903 Numbers 8, 9; August, 1904 Numbers 4, 7, 8, 9, 10; April, July, August, 1905 Numbers 4, 9; April, September, 1906 Numbers 5, 10, 11; May, October, November, 1907 Numbers 5, 9; March, May, September, 1908 Numbers 11; April, November, 1909 Numbers 1911 10; July, October, Numbers 8, 11; March, August, November, 1912 Numbers , 2; January, February, 1913 1 Numbers 1, 3, 4; July, September, October, 1913 7 Numbers , 10; July, October, 1916 Number 10; October, 1919 Number 11; November, 1925 Volume Volume Volume Volume Volume Volume 19912.2 Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume 59, 61, 63, 68, 69, 94 Numbers 3, 8, 11; March, August, November, 1925 Number 5; May, 1927 Number 8; August. 1929 Number 4; April, 1936 Number September, 1937 Number March, 1938 5, Number 1; January, 1895 7. 9, Number 2; 11, 13, 15, 16, 17, 20, 21, 23, 25, 26, 27, 28, 1906 29, 32, 33, Number 11; March, 1896 January, 1897 Number 22; June, 1898 Number 20, May, 1899 Number 26 June, 1900 ‘0 Number 1; July, 1900 Number 24; June, 1901 Numbers 10, 11; September, 1902 Number 9; February, 1903 Number 26; June, 1904 Numbers 5, 15; February, April, 1905 Numbers 15, 16; October, 1905 Numbers 2, 12, 13; January, March, 1906 Numbers 4, 11, 25, 26; July, September, December, Number 18; May, 1907 Number 5; July, 1908 Numbers 7, 15, 24; February, April, June, 1909 Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume Volume 37, 38, 39, 41, 42, 43, 49, 51, 54, 56, 64, 66, 70, 73, 79, 81, 86, 87, 91, 93, Number Number Number Number Number Number Number Number Number Number Number Number Number Number Number Numbers 1, Number Number Number Number 95 2; January, 1911 1; July, 1911 10; May, 1912 6; March, 1913 7; October, 1913 9; May, 1914 5; March, 1917 2, 9; January, May, 1918 12; December, 1919 12; December, 1920 2; August, 1924 3; December, 1925 3; December, 1927 6; June, 1929 6; June, 1932 2; January, February, 1933 ; November, 1935 ; January, 1936 February, 1938 l-‘Nl-‘ko ‘0 January, 1939 "I7'1111'1111'1111111'75