Ln lWWllW ”WW WW WWW 31293 00860 61843 ’ Michigan Scam University This is to certify that the thesis entitled GENETIC STUDIES OF TASTE PERCEPTIONS 0F ANT IDESMA AND PHENYLTHIOCARBAMIDE presented by Frankie Johnson Brown has been accepted towards fulfillment of the requirements for 911- DL degree in 19.le— 17' Major profes£r(J Date &/&7/X/ 0-7639 ovsnou FINES: 25¢ PIP day nor its- Rnynuguc' LIBRA'RY MATERIALS; Place in book return to move charge from circulation records . IFS '0»! a is I I -4, ' "::\\\ ' ~ 3 (fig 3‘: *- “ ~23 1W IL l ‘\"I” I .4 @ Copyright by FRANKIE JOHNSON BROWN 1981 GENETIC STUDIES OF TASTE PERCEPTIONS OF ANTIDESMA AND PHENYLTHIOCARBAMIDE By Frankie Johnson Brown A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Zoology 1981 ABSTRACT GENETIC STUDIES OF TASTE PERCEPTIONS OF ANTIDESMA AND PHENYLTHIOCARBAMIDE By Frankie Johnson Brown To investigate the reported association between bitter taste responses to phenylthiocarbamide (PTC) and aqueous extracts of the tropical fruit, Antidesma bunius by Henkin and Gillis, (1977), taste perceptions of 968 unrelated individuals and 470 related subjects (115 families and 12 twin pairs) were assessed for three PTC concentrations (20.31, 40.63 and 81.25 mg/l), two preparations of Antidesma (aqueous extract and liquified macerated materials) and six control solutions (lemon juice, distilled water, 1M NaCl, 0.001M quinine sulfate, 0.5M sucrose and grape juice). Frequencies of specific taste responses for each solution recorded by all subjects were analyzed by age, sex, race/ethnic group, smoking status and elapsed time since last fbod eaten and comparisons of PTC and Antidesma perceptions were made. Taste perceptions of PTC and Antidesma obtained from families and twins were additionally analyzed to determine if these re5ponses were consistent with a simple dominant-recessive genetic hypothesis. Perceptual errors made by subjects in the identification of controls for sweet, tasteless, salty, sour and bitter taste qualities Frankie Johnson Brown were less than those reported by previous studies. There were no significant associations of these errors with age, sex, race, smoking status nor elapsed time since food ingestion. Misidentification of controls did not appear to produce significant differences in taste responses to PTC nor Antidesma. Based on responses to the PTC concentration of 81.25 mg/l, taster and nontaster frequencies were 75.8 percent and 24.2 percent. Corresponding bitter and nonbitter responses were 70.1 percent and 29.9 percent. These frequencies and those obtained for responses to the concentrations of 20.31 and 40.63 mg/l did not appear significantly affected when age groupings were compared. PTC perceptions however, did appear to be affected by sex, race, smoking status and elapsed time since last food eaten. Diversity of taste perceptions of Antidesma were observed both for the Aqueous extract (Antidesma I) and for liquified macerated materials (Antidesma II). Major perceptions of Antidesma I were sweet (50.9%) and sour (36.0%) and for Antidesma II, sour (45.9%) and bitter (28.2%). The finding of significant differences between the overall responses and for the dichotomous classifications of the major percep- tions of these solutions suggested inherent compositional variations. The Antidesma perceptions did not appear to be significantly affected by smoking nor elapsed time since food ingestion. However, effects of age, sex and race were suggested. Specific taste perceptions of Antidesma I were not significantly associated with any taste response for each of the PTC concentrations nor with PTC tasting status. Conversely, overall perceptions, as well Frankie Johnson Brown as bitter-nonbitter perceptions of Antidesma II, showed significant correlations with PTC responses, primarily due to the less than expected frequency of subjects who judged both Antidesma II and PTC as bitter. Contrary to the Henkin and Gillis report however, no mutual exclusivity of bitter perceptions for either Antidesma II or I and PTC were observed. Results from comparisons of observed and expected PTC taster- nontaster progeny frequencies from various mating types were in excellent agreement with the well established dominant-recessive hypothesis. Support for this genetic hypothesis for Antidesma taste perceptions in families was found only in the case of bitter-nonbitter responses for Antidesma I. Comparisons of twin concordance rates for Antidesma perceptions revealed no significant differences between concordance of M2 and 02 twins. Similar results were obtained when MZ-DZ twin concordance rates for PTC responses were compared. However, definitive conclusions were unwarranted due to the small sample size of twins studied. To C. B. and Chuckie and my parents, William P. and Bernice M. Johnson iii ACKNOWLEDGMENTS I wish to express my sincere appreciation to Dr. James V. Higgins, my major professor, for his guidance and counsel throughout my graduate experience which has now extended through the Master's and Doctorial programs. I also wish to thank Drs. John Shaver, Emanuel Hackel and Allan Morris for serving on my graduate committee and pro- viding constructive criticisms and advice. A special expression of thanks is given to Mrs. Catherine Sweeney of the Kampong, Coconut Grove, Florida for graciously supplying the Antidesma fruit which was used in the study and to the late Dr. William Gillis for his valuable suggestions in the initial planning of the project. I am also grateful to the entire faculty and staff of the Department of Natural Science for their support, cooperation and actual participation and for allowing me to request volunteers from their students as subjects in this study. I am especially indebted to Dr. Ben Cathey for his numerous contributions of support, encouragement and technical assistance throughout this project, Dr. Alain Corcos for his advocacy and advice and Dr. Don Weinshank, for his invaluable help with computer processing and analysis of data. Special appreciation for statistical advice and/or computer programing is extended to Drs. Forrest Carter, John Gill and Joe Byers, iv and thanks goes to Drs. James Butcher and Jerry Cash for their unique support and advice. Additionally, much appreciation is given to Annetta Brock, Dan O'Malley, Sharon Cardwell, Montios Chavis and Brenda Mills for their special assistance, friendship and moral support. ‘This document could not have been completed without the cooperation of students and staff at Michigan State University and families of the East Lansing area who willingly participated in the study. I am especially appreciative for the constant encouragement, gentle prodding and expressions of confidence provided by my parents, William and Bernice Johnson and sibs, Mary Alice, Willa, Pete and Charles. Most of all, I shall be eternally grateful to my husband C. B. and son Chuckie for their patient understanding, continuous reassurance and sustaining support throughout my entire graduate program. LIST OF TABLES . . LIST OF FIGURES . . INTRODUCTION . . . LITERATURE REVIEW . MATERIALS AND METHODS TABLE OF CONTENTS Sampling Procedures . . . . . . . . . . . . Testing Procedures . Taste Solutions Preparation and Processing . . . . RESULTS . . O O O O O O O I O O O O O O O 0 Overall Taste Perception Frequencies and General Demographic Data 0 O O O O I O O O O O O 0 Taste Perceptions Taste Perceptions Taste Perceptions Taste Perceptions Taste Perceptions and Age of Respondents . . . . . . and Sex of Respondents . . . . . . and Race of Respondents . . . . . and Smoking Status of Respondents and Time of Last Food Eaten (Elapsed Time) By Respondents . . . . . . . . . . Antidesma Perceptions and PTC Responses . . . . . . Comparisons of Taste Perceptions of Antidesma I and Antidesma II . . . . . . . . . . . . . . . . . Taste Perception Family Studies . . . . . . . . . . Genetic Analyses of Antidesma and PTC Taste Perceptions . DISCUSSION . Taste Perceptions of Controls . . . . . . . . . . . Taste Perceptions of PTC . . . . . . . . Taste Perceptions of Antidesma . . . . . . . . . . . Antidesma Perceptions and PTC Responses . . . . . . Family Studies of Taste Perceptions . . . . . . . . Genetic Analysis of PTC and Antidesma Taste Perceptions Summary 0 O O O O O O O O O O O I O 0 O 0 vi Page viii xiii 22 22 23 24 27 27 36 50 S7 66 71 81 91 105 110 145 146 150 156 161 163 163 166 APPENDIC APPENDIX A. B. C. D. E. BS 0 O O O O O O O O O O O O O O O O Derivation of Snyder's Ratios . . . Comparison of PTC and Antidesma Responders and Nonresponders (Henkin and Gillis, 1977) Letters of Introduction to Households Consent Form and Survey Questionnaire Rationale for Use of Different Statistics Employed . BIBLIOGRAPHY . . . . . . . . . . . . . . . . vii O Page 171 171 172 173 176 179 188 Table 1a. 1b. 2a. 2b. 3a. 3b. 4a. 4b. 5a. Sb. Sc. 6a. 6b. 7a. 7b. LIST OF TABLES Taste Perceptions of Controls . . . . . . . . . Intensities of Controls . . . . . . . . . . . . . Taste Perceptions of Antidesma . . . . . . . . . Intensities of Antidesma . . . . . . . . . . . . . Comparison of Antidesma I Perceptions with Mis- classifications (Errors) of Controls . . . . . . . Comparison of Antidesma II Perceptions with Mis- classifications (Errors) of Controls . . . . . Taste Perceptions of PTC . . . . . . . . . . . . . Intensities of PTC . . . . . . . . . . . . . . . . Comparison of Taste Perceptions of PTC (Low Con- centration) with Misclassifications (Errors) of Controls . . . . . . . . . . . . . . . . . . Comparison of Taste Perceptions of PTC (Medium Concentration) with Misclassifications (Errors) of Controls . . . . . . . . . . . . . . . . . . Comparison of Taste Perceptions of PTC (High Con- centration) with Misclassifications (Errors) of Controls . . . . . . . . . . . . . . . . . Comparison of Age of Respondent with Perception of Controls . . . . . . . . . . . . . . . . . Comparison of Age of Respondent and Misclassifi- cations (Errors) of Controls . . . . . . . . . . Comparison of Age of Respondent with Perception Of Mt idesma O O O O O O O I O O O O O O O O O 0 Comparison of Age of Respondent with Antidesma Bitter-Nonbitter Perceptions . . . . . . . . . . viii Page 29 29 31 31 33 34 35 35 37 38 39 41 42 43 45 Table Page 8a. Comparison of Age of Respondent with Perception Of PTC O O O O O O O O O O O O O O O O O O O O O O I O O 46 8b. Comparison of Age of Respondent with PTC Taster-Nontaster and Bitter-Nonbitter Perceptions . . . . . . . . . . . . . . . . . . . . . . 49 9a. Comparison of Sex of Respondent and Perception of Controls . . . . . . . . . . . . . . . . . . . . . . 51 9b. Comparison of Sex of Respondent with Misclassifi- cation of Controls . . . . . . . . . . . . . . . . . . . 52 10a. Comparison of Sex of Respondent with Perceptions Of Mtidesma O O O I O O O O O O O I O O O O O O O O O O 54 10b. Comparison of Sex of Respondent with Antidesma Bitter-Nonbitter Status . . . . . . . . . . . . . . . . 54 lla. Comparison of Sex of Respondent and Perception Of per I O I O O O O O O O O O O O O O O O O O O 0 O O 0 55 11b. Comparison of Sex of Respondent and PTC Taster- Nontaster and Bitter-Nonbitter Status . . . . . . . . . 56 12a. Comparison of Race of Respondent and Perceptions of Controls . . . . . . . . . . . . . . . . . . . . . . 58 12b. Comparison of Race of Respondent and Misclassifi- cation (Errors) of Controls . . . . . . . . . . . . . . 59 13a. Comparison of Race of Respondent with Perceptions of Antidesma . . . . . . . . . . . . . . . . . . . . . . 61 13b. Comparison of Race of Respondent with Bitter- Nonbitter Antidesma Perceptions . . . . . . . . . . . . 63 I43. Comparison of Race of Respondent with Perception Of PTC O O O O O O O O O O O O I O I O O O O O 0 O O O O 64 14b. Comparison of Race of Respondent and PTC Taster- Nontaster and Bitter-Nonbitter Status . . . . . . . . . 65 15a. Comparison of Smoking Status of Respondent and Perception of Controls . . .7. . . . . . . . . . . . . . 67 15b. Comparison of Smoking Status of Respondent and Misclassification of Controls . . . . . . . . . . . . . 68 ix Table Page 16a. Comparison of Smoking Status of Respondent with Perceptions of Antidesma . . . . . . . . . . . . . . . . 69 16b. Comparison of Smoking Status of Respondent and Antidesma Bitter-Nonbitter Status . . . . . . . . . . . 71 17a. Comparison of Smoking Status of Respondent with Perception of PTC . . . . . . . . . . . . . . . . . . . 72 17b. Comparison of Smoking Status of Respondent and PTC Taster-Nontaster and Bitter-Nonbitter St atus O O O O O O O O O O O O O O O O O O O O O O O O O 73 18a. Comparison of Elapsed Time (Time of Last Food Eaten by Respondent) and Perception of Controls . . . . 74 18b. Comparison of Elapsed Time (Time of Last Food Eaten by Respondent) and Misclassification (Errors) of Controls . . . . . . . . . . . . . . . . . . 75 19a. Comparison of Elapsed Time (Time of Last Food Eaten by Respondent) with Perceptions of Mtidesma O O I O O O O O O O O O O O O O O O 0 O O O O 77 19b. Comparison of Elapsed Time (Time of Last Food Eaten by Respondent) with Antidesma Bitter- Nonbitter Status . . . . . . . . . . . . . . . . . . . . 78 20a. Comparison of Elapsed Time (Time of Last Food Eaten by Respondent) and Perceptions of PTC . . . . . . 79 20b. Comparison of Elapsed Time (Time of Last Food Eaten by Respondent) and PTC Taster-Nontaster and Bitter-Nonbitter Status . . . . . . . . . . . . . . 80 20c. Comparison of PTC Taster-Nontaster and Bitter- Nonbitter Status with Elapsed Time: Less than One Hour Versus Greater than One Hour . . . . . . . . . 82 21a. Comparison of Taste Responses of Low Concentration of PTC and Perceptions of Antidesma . . . . . . . . . . 83 21b. Comparison of Antidesma I and II Responses with PTC Low Taster-Nontaster and Bitter-Nonbitter St atus O O O O O O I I I O O O O O O O O O I O O O O O O 85 21c. Comparison of Antidesma Bitter-Nonbitter Responses with PTC Low Taster-Nontaster and Bitter-Nonbitter St atus C C C O O O O C O I C O O O O O O I O C I I O O O 87 Table Page 22a. Comparison of Taste Responses of Medium Concentra- tion of PTC with Perceptions of Antidesma . . . . . . . 88 22b. Comparison of Antidesma I and II Responses with FTC Medium Taster-Nontaster and Bitter- Nonbitter Status . . . . . . . . . . . . . . . . . . . . 89 22c. Comparison of Antidesma Bitter-Nonbitter Responses with PTC Medium Taster-Nontaster and Bitter- Nonbitter Stattls I I I I I I I I I I I I I I I I I I I I 92 23a. Comparison of Taste Responses of High Concentration of PTC with Perceptions of Antidesma . . . . . . . . . . 93 23b. Comparison of Antidesma I and II Responses with PTC High Taster-Nontaster and Bitter-Nonbitter Status I I I I I I I I I I I I I I I I I I I I I I I I I 94 23c. Comparison of Antidesma Bitter-Nonbitter Responses with PTC High Taster-Nontaster and Bitter- Nonbitter Status I I I I I I I I I I I I I I I I I I I I 96 24a. Comparison of Taste Perceptions of Antidesma I and Mtidesma II I I I I I I I I I I I I I I I I I I I I 97 24b. Comparison of Taste Perceptions of Antidesma Solu- tions: Overall Antidesma Perceptions with Antidesma II (Bitter-Nonbitter, Sour-Nonsour, Sweet-Nonsweet) Perceptions . . . . . . . . . . . . . . 100 24c. Comparison of Antidesma I (Bitter-Nonbitter, Sour-Nonsour, Sweet-Nonsweet) Perceptions with Antidesma II (Bitter-Nonbitter, Sour-Nonsour, Sweet-Nonsweet) Perceptions . . . . . . . . . . . . . . 101 25a. Comparison of Parental and Offspring Perceptions of Controls . . . . . . . . . . . . . . . . . . . . . . 105 25b. Comparison of Misclassification (Errors) of Controls for Parents and OffSpring . . . . . . . . . . . 106 26a. Comparison of Taste Perceptions of Antidesma for Parents and Offspring . . . . . . . . . . . . . . . . . 107 26b. Comparison of Antidesma Bitter-Nonbitter Responses fer Parents and Offspring . . . . . . . . . . . . . . . 108 27a. Comparison of Taste Perceptions of PTC for Parents and Offspring . . . . . . . . . . . . . . . . . . . . . 109 Table Page 27b. Comparison of Taster-Nontaster and Bitter-Nonbitter Perceptions of PTC fbr Parents and Offspring . . . . . . 126 28. Family Studies: PTC Taster-Nontaster Perceptions . . . . . 127 29. Family Studies: Antidesma I Bitter-Nonbitter Per- cept ions I I I I I I I I I I I I I I I I I I I I I I I I 130 30. Family Studies: Antidesma I Sweet-Nonsweet Per- ceptions . . . . . . . . . . . . . . . . . . . . . . . . 133 31. Family Studies: Antidesma I Sour-Nonsour Perceptions . . . 134 32. Family Studies: Antidesma II Bitter-Nonbitter Per- ceptions I I I I I I I I I I I I I I I I I I I I I I I I 1 35 33. Family Studies: Antidesma II Sweet-Nonsweet Per- ceptions I I I I I I I I I I I I I I I I I I I I I I I I 1 36 34. Family Studies: Antidesma II Sour-Nonsour Percep- tions . . . . . . . . . . . . . . . . . . . . . . . . . 137 35. Taste Perceptions of PTC and Antidesma of Twins . . . . . 140 36a. PTC Taste Perceptions of Twins: Concordance Rates . . . . 142 36b. Antidesma Taste Perceptions of Twins: Concordance Rates . . . . . . . . . . . . . . . . . . . . . . . . . 144 37. Population Frequencies of PTC Nontasters . . . . . . . . . 154 38. Comparison of Antidesma Responders and Nonresponders . . . 159 LIST OF FIGURES Antidesma bunius (bignai) . . . . . . . . . . Tree of Antidesma bunius . . . . . . . . . . . . . Fruits of Antidesma bunius . . . . . . . . . . . . Antidesma and PTC Taste Responses . . . . . . Genetic Studies of Taste Perceptions of Antidesma and Phenylthiocarbamide—-Fami1y Pedigrees . xiii Page 16 l7 18 20 112 INTRODUCTION Genetic differences in taste responsiveness were first demon- strated fbr phenylthiocarbamide (PTC or phenylthiourea) and related compounds. Such taste perception differences are assumed to be dependent upon the presence of the N-C=S radical of these compounds which is typically perceived as bitter or tasteless although other taste qualities have been reported. Since the initial descriptions, numerous population studies have confirmed the divergent taste perceptions for PTC and have resulted in the classification of individuals as tasters or nontasters. The frequency of tasters has been fbund to be approxi- mately 30 percent in American Caucasian populations but varies from 0-49 percent in other racial groups (Corcos and Scarborough, 1978). Additionally, specific threshold concentration effects have been observed which appear to increase with age and are generally reported decreased in females. Differential frequencies also appear to be associ- ated with certain types of disease entities especially those which relate to thyroid functioning. Furthermore correlations of PTC taste sensitivity have been reported fbr a variety of other substances. Taste perceptions of PTC are generally considered to be con— trolled by a single pair of alleles. The ability to taste PTC is thought to be inherited as a Mendelian dominant while the inability to taste this substance is due to homozygosity fbr the recessive allele. This hypothesis has been largely confirmed although occasional incomplete penetrance of the "taster" allele has been reported and a multiple allelic system has been postulated to account for extremely sensitive tasters in certain populations (Das, 1956; Lugg, 1970). Recently, Henkin and Gillis (1977) confirmed divergent taste responses to extracts from berries of the Antidesma bunius tree. Their investigation was initiated in response to a prior incident in which two of eight persons served a pie made from Antidesma berries com- plained that the pie was extremely bitter and inedible while the other six persons found the pie pleasant tasting and sweet. These observa- tions were considered unusual since it was known that Antidesma fruit has been extensively used as food by natives of South East Asia and Florida and fbr many years has been eaten in pies, jams, jellies and sauces or as raw fruit. In the study of 170 subjects, Henkin and Gillis not only found differences in taste perceptions to extracts prepared from Antidesma fruit but also concluded that these differences were specifically related to taste perceptions of PTC. In their study, responders to PTC and Antidesma extract were defined as those who described these solu- tions as bitter while non-responders were defined as those who judged the solutions as either tasteless or of another taste quality. Among the bitter responders to PTC, there were no bitter responders to Antidesma and conversely, among the bitter responders to Antidesma, there were no bitter responders to PTC. Based on these observations, these researchers concluded that some type of interaction may exist between those factors which are responsible for bitter cognition of these two substances since no single individual sampled perceived both of these as bitter. Furthermore they suggest that the relationship of these factors may occur on a functional or a genetic level. In view of these findings the present study was proposed to sample larger numbers of subjects to establish frequencies of different taste responses of Antidesma, to further investigate the associations between Antidesma and PTC perceptions and to conduct family studies to determine if divergent taste responses to Antidesma confbrm to a simple genetic hypothesis. LITERATURE REVIEW Among the numerous attributes used to describe diversity in human populations are those involving variations in drug sensitivity. One of the most widely investigated of these is taste perceptions of phenylthiocarbamide (PTC or phenylthiourea) which were first described in 1931. A. L. Fox (1931) who had synthesized this substance received complaints from colleagues saying that the laboratory air contained an intensely bitter dust. Fox and some of his other colleagues, however, did not perceive this bitter sensation and when he placed PTC crystals on the tongues of various individuals he found that some experienced the intensely bitter taste while others found the crystals tasteless. Since this initial description, investigations conducted in numerous populations have confirmed the taster-nontaster dichotomy. PTC is a synthetic organic compound, belonging to a group of chemically related substances commonly regarded as goitrogens because of their anti-thyroid activity. Over 100 of these compounds both naturally occurring and synthetic are now known and they are related in chemical structure by the presence of a N-C=S group. This common chemical grouping has been shown to be responsible for the bitter per- ceptions of sensitive individuals (Fox, 1932; Hopkins, 1942; Harris and Kalmus, 1950; Barnicot gt 31., 1951). Not all persons find the taste of PTC bitter or neutral. Blakeslee and Fox (1932) and Blakeslee (1935) as well as others have reported that some people find PTC sweet and others find it salty, sour, camphory or sulfury. Skude (1959, 1960a) reported that about 7-9 percent of his population tested found PTC sweet tasting and he considered this to be an inherited characteristic. Later with repeated testing of these subjects he found considerable variation in responses and thus suggested that further study was required before definitive conclusions could be drawn (Skude, 1960b). The finding of these "deviant" PTC taste responses has however, led to inconsistencies in classifications by some workers. In some studies individuals are classified as tasters regardless of which type of taste quality is per- ceived while other studies record tasters as only those who judge PTC as bitter. A variety of methods have been used to study the PTC tasting phenomenon. In earlier studies, PTC crystals were placed directly on the tongue. Later it became popular to impregnate filter papers with certain concentrations of PTC, let the papers dry then have individuals chew on the paper beginning with the lowest concentrations and proceed- ing to higher concentrations. By this method, thresholds of sensitive individuals could be determined and typically resulted in a bimodal distribution in which the antimode was taken as the dividing line between tasters and nontasters. Another technique used to determine thresholds was introduced by Blakeslee (1932) and refined by Hartman (1939) involved taste testing by use of varying concentrations of PTC in solutions. This latter method was thought to result in a lower percentage of'misclassifications of the three techniques. A revised version of the solution method as employed by Harris and Kalmus (1949), has been used fbr most population studies with minor variations by several researchers. In the majority of these studies the solution con- taining 81.25 mg/l has been used to separate tasters from nontasters. When these methods were used to determine PTC threshold per- teptions, population differences with respect to sex have been noted. While the absolute proportions of tasters and nontasters appear to be of equal frequency in males and females, several studies have con- cluded that on the average, female tasters can detect PTC in higher dilutions (Hartman, 1939; Falconer, 1946; Mohr, 1951; Montenegro, 1964). In some studies, the sex differences have been highly significant while others report only slight differences in sensitivity between the sexes (Than-Than-Siht E£.El" 1974). An additional variation in threshold sensitivity has been observed with age. Harris and Kalmus (1949) in a study of 441 British males, found that the modes of the taster and nontaster groups, as well as the antimode dividing the two groups were shifted in the direction of the more concentrated solutions with increasing age. They concluded that a deterioration of taste sensitivity of about one dilution step occurs for each additional twenty years of age. Although less drastic changes have been noted by other researchers (Mohr, 1951), it is gene- rally agreed that PTC sensitivity decreases with age. Shortly after the initial descriptions of divergent taste responses to PTC, independent studies conducted by Blakeslee (1932) of 103 families and Snyder (1932) of 800 families concluded that this taste sensitivity was an inherited phenomenon which was determined by a single pair of alleles. Furthermore, it was suggested that the ability to taste PTC was determined by a Mendelian dominant while nontasters were homozygous for the recessive allele. In the development of these hypotheses, Snyder (1932) formulated his now classic ratios for testing data from family studies for a simple dominant-recessive mode of inheri- tance. His initial assumption that nontasting is recessive was based on the observation that matings of nontaster parents produced almost exclusively nontaster offspring while matings of tasters produced both taster and nontaster progeny. Thus tasters would be homozygous or heterozygous for the dominant taster allele. Then assuming Hardy- Weinberg conditions in which the frequency of the homozygous dominant (taster) = p2, heterozygotes (tasters) = 2pq and homozygous recessives (nontasters) = q2, Snyder concluded that it was possible to predict the percentages of recessive offspring expected from various matings of parents displaying the dominant or recessive trait using the following formulae (fbr derivations, see Appendix A): Percent Recessives From Dominant x Recessive Matings: and Percent Recessives From Dominant x Dominant Matings: 2 s2 =(—1-J;—q—)-z (Snyder, 1932:Modified). Analysis of PTC pedigree data by use of these ratios produced close agreement of expected and observed frequencies, thus it was assumed that a single pair of alleles was responsible for the inheri- tance of taste reactions to PTC. Despite the apparent "goodness of fit" for Snyder's genetic model, others have proposed modifications of this simple dominant— recessive inheritance pattern fbr PTC perceptions. In an analysis of 845 sibling pairs, Das (1956) concluded that his results could be best explained by modifying the monogenic theory to assume 90 percent pene- trance of the dominant allele. Additional support for reduced penetrance as well as variable expressivity of the taster allele has been provided by the finding that some people are able to detect the bitter taste at very high dilutions while others detect it only with crystals. (Some individuals have also been discovered who are unable to taste extremely high solution concen- trations nor the PTC crystals.) Furthermore, studies by Lugg (1966, 1968, 1970) have suggested that a multiple allelic hypothesis is neces- sary to account for the multimodal threshold distributions obtained in his study of population groups containing individuals with unusually high PTC taste acuity. Similar conclusions have been reached by Rychkov and Borodina (1973), from extended investigations of PTC hypersensitiv- ity from which they proposed triallelic autosomal control of PTC sensi- tivity. Other researchers have suggested a polygenic inheritance mode. These hypotheses however have not been supported by others. Indeed, extensive studies by Rao and Morton (1977) of PTC taste sensitivity in a large sample from Brazil (2,090 parents and 2,245 offspring) and sub- sequent application of a mixed model of complex segregation analysis have found no evidence for incomplete dominance, polygenic variation, nor did they suggest any effect of family environment on PTC sensitivity. They concluded that skepticism about simple recessivity is unwarranted. Numerous population studies have revealed considerable differ- ences in the proportion of tasters and nontasters in different parts of the world. These studies have been important for anthropological reasons to suggest possible ethnological factors involved in PTC sensitivity. Among the Caucasian p0pulations of Western Europe and North American origin, the frequency of nontasters is approximately 25-35 percent (Allison and Blumberg, 1959), among American Negroes, 8-20 percent (Johnston gt 31., 1966; Lee, B. F., 1934), among African Blacks, 3-12.5 percent (Barnicot, 1950; Scott-Emuakpor 33 al., 1975), among Chinese, 6-10.6 percent (Cohen and Ogden, 1949; Barnicot, 1950), among American Indians, 6 percent (Cohen and Ogden, 1949). In general, it appears that Negroid, Mongoloid and American Indian populations are characterized by a lower percentage of nontasters (less than 20 percent) while Caucasian populations typically contain 25—35 percent nontasters. The highest nontaster frequencies (greater than 50 percent) have been reported for certain Australian aborigines and some groups in India (Basu and Ghost, 1968). The nontaster frequencies for other groups may vary from 0-49 percent depending on geographical origin and racial composition (Corcos and Scarborough, 1978; Garr, 1934). Appearance of the PTC taste divergence dates back to prehuman times. Fischer 35 21° (1939) in a study of chimpanzees in Great Britain zoos found a frequency of 26 percent nontasters. Corresponding values fbr nontasters in their human population studies were 25-30 percent. From these observations, it was concluded that such consistency between human and anthropoid groups is attributable to "a stably balanced and enduring dimorphism that has kept the ratio the same over millions of generations since the separation of anthropoid and humanoid stock." 10 The maintenance of the PTC taste polymorphism has been the sub- ject of much speculation. As with other polymorphisms, it is believed that this kind of biochemical diversity can only be maintained by a balance of selective fOrces acting on the various phenotypes. Several theories concerning the possible selective advantage of both tasters and nontasters have been postulated. Basic to these hypotheses have been the observations of differential frequencies of taste sensitivities associated with various human conditions. The fbllowing represents an enumeration of some of the more widely studied associations. A significant increase in taster phenotypes has been reported to be correlated with dental caries in adults under the age of 40—50 (Tibera-Dumitru, 1965), malignant tumors of the ovaries, uterus and breasts in females (Milunicova gt 31., 1969), inflammatory diseases (such as rheumatoid arthritis and ankylosing spondylitis, Stepan gt 31., 1965), greater maturation in visual-motor perception (Greene, 1974), increased skeletal maturity (Johnston £3 31., 1966), and tuberculosis (Saldanha, 1956). Conversely, the proportion of nontasters is reputedly increased in primary glaucoma diseases (Becker and Morton, 1964) and ' diabetes mellitus (Terry, 1950; Rao and Sisodia, 1970). It should be noted that the above associations with tasting status have not been universally confirmed by subsequent studies but merely suggest possible mechanisms by which various taster alleles may be maintained in popu- lations (Kalmus and Lewkonia, 1973; Lasker and Fernandez, 1970). Perhaps the most widely studied relationships linked to the PTC polymorphism have been those involving thyroid functioning, some of which have been alluded to earlier. Investigations of this association have been numerous because of the well known goitrogenic effects of PTC 11 related compounds. Furthermore, a number of these related antithyroid compounds are present in small amounts in many edible plants of the Brassica genus including cabbage, kales, brussel sprouts, turnips, etc. (Boyd, 1950; Van Etten, 1969). Since PTC itself does not occur in nature, what is seen as the PTC taste polymorphism has been thought to reflect individual ability to detect and perhaps reject a large number of naturally occurring goitrogens. Studies by Greene (1974) on iodized and noniodized populations in which goiter is endemic in areas where a number of PTC like goitrogen containing plants are consumed in moderate quantities, have found significant correlations between PTC taste sensi- tivity and visual-motor maturation and an increase in taste sensitivity with age in the noniodized individuals but not in those which were iodized. From these findings the author concludes that sensitive tasters of PTC may limit their ingestion of the bitter tasting goitrogens, reduce the stress placed on their thyroid gland and thus increase the likeli- hood of normal neurological maturation under these particular environ- mental conditions. Several other reports have linked the ingestion of plant pro- duced goitrogens with endemic goiter (Clements and Wishart, 1956; Greene .EE.El°» 1958; Peltola, 1960; Barzelatto and Covarrubias, 1969). Addi- tional studies have confirmed the linkage of PTC taste sensitivity to goiter, both sporadic (Harris gt 31., 1949; Kitchin 33 21., 1959) and endemic (Brand, 1963; Azevedo £5 31., 1965). Most of these studies have concluded that nontasters show a significantly increased prevalence of nodular as opposed to diffuse goiter (Mendez 33 31., 1972; Boyce gt ‘21., 1976). Furthermore, other investigations have fbund a significant excess of nontasters among athyreotic cretins as well as a similar 12 increased nontaster frequency among the parents and siblings of the cretins (Shepard and Gartler, 1960; Shepard, 1960; Fraser, 1961). These researchers suggest that "the nontaster fetus may be more suscep- tible to embryonic thyroidectomy by naturally occurring goitrogens in the diet of the mother." Such conclusions have led to the hypothesis that tasters are at a selective advantage over nontasters under environ- mental conditions where iodine intake may be low and naturally occurring goitrogens are consumed in significant quantities. Under different conditions selection may favor the nontaster phenotype. Evidence for this assumption has been suggested by the significantly lower prevalence of hyperthyroidism (toxic goiter) among nontasters (Kitchin 35 31., 1959; Persson gt 31., 1972). In fact, Farid £5 31. (1977) have suggested that tasters who also possess the HLA B-8 antigen have a 5-8 fold increased risk of developing Graves disease (a fbrm of hyperthyroidism). Additionally, Milunicova gt 31. (1969) have demonstrated a significantly lower incidence of carcinoma of the thyroid among women who are nontasters. From the foregoing, it has been assumed by several researchers that the tasting polymorphism has probably been maintained due to selection against the two homozygote genotypes under different condi- tions and perhaps even at different points in the life cycle, thus pro- ducing relative heterozygote advantage (Greene, 1974). What is unclear however, is the mechanism of action of the taster alleles. Whether they simply represent a pleiotropic expression of genes coding for thyroid function or merely those responsible for some variation in the rejec- tion mechanism or disposal of antithyroid substances remains unknown (Fraser, 1961; Kalmus, 1972). At present there exists no satisfactory 13 evidence to prove a causal relationship between PTC tasting status and the occurrence of both thyroid and nonthyroid related conditions in human populations. Of the more interesting nonpathologic associations of taster status of PTC, have been those involving relationships with taste per- ceptions of other substances. As indicated earlier, most of these corre- lations have been established for the more than 100 PTC related com- pounds which contain the N—C=S group. Such substances show threshold taste distributions in populations similar to those of PTC. A small number of investigations however, have been undertaken which suggest associations of PTC perceptions and other non-PTC like compounds. It is of interest that most of these compounds have been those which elicit bitter perceptions to most individuals and have been studied presumably to assist in elucidating the physiological nature as well as number and types of receptors responsible for bitter cognition in humans. Fischer and Griffin (1964) have reported that the degree of sensitivity for quinine, influences the expression of taste sensitivity for PTC-type compounds such as 6-N-propy1thiouracil (PROP). Their data showed that the average PROP taste threshold for each of the tasting and nontasting modes is significantly higher for very insensitive tasters of quinine than fer sensitive tasters. These workers suggest that the influence of quinine taste sensitivity on the expression of PROP responsiveness may be regarded as an example of partial epistatis in humans. A more recent study by Bartoshuk (1979) suggested that the intensity of the bitter taste of saccharin is also related to the taste sensitivity to PROP. Based on an analysis of scaled intensities of the sweet, salty, sour and bitter taste qualities of sodium saccharin by tasters and 14 nontasters of PROP, it was concluded that saccharin tastes significantly less bitter to nontasters at the concentrations used in popular diet beverages. Hall 33 31. (1975) have examined the relationship between PTC taste perception and the taste of caffeine. Their assessment of taste thresholds for PTC and caffeine produced a bimodal distribution fer both of these compounds. The bimodality of caffeine thresholds however, was restricted to the lower concentrations but was highly corre- lated to PTC thresholds. Thus these workers concluded that sensitivity to the taste of PTC predicts sensitivity to caffeine. Another apparent relationship to PTC perceptions which has formed the basis fbr the present study were the findings of Henkin and Gillis (1977) which linked specific PTC perceptions to aqueous extracts from berries of the Antidesma bunius tree. Since infbrmation about this fruit is not widely disseminated, a brief description of the plant as well as the findings of these workers fbllows. Antidesma bunius is a member of a large genus of dioecious shrubs and small trees of the family Euphorbiaceae native to tropical Asia, Africa, Australia and the Pacific, particularly in the islands of the Phillipines, Indonesia and the Malay Peninsula (Burkill, 1935; Benthall, 1946). In these areas the plant is referred to by a variety of common names depending on the area in which it grows (e.g., Bignay in the Phillipines, Booni in Malay, Boorneh in West Java, etc., Fairchild, 1939). Introduction of the fruit in this country appears to have occurred around 1913 according to a U.S. Department of Agriculture Report and since that time, has been grown exclusively in South Florida, specifically in the Fairchild Tropical Gardens near Coconut Grove, Florida (Fairchild, 1939; Sturrock and Menninger, 1946). The fruit 15 grows in large clusters like grapes (Fig. l, 2 and 3) although each fruit is about the size and color of a blueberry when ripe and is typically described as ovoid, fleshy and sub-acid, each containing a single seed. The fruiting season varies in different parts of the world but in this country, fruits are commonly found from late summer to early winter. The scientific name Antidesma was given to the tree to denote its use by natives of Ceylon as a cure for snake bite, according to the Dutch botanist, J. Burmann (1737). According to Burkill (1935), the bark is poisonous, containing an alkaloid but is used medicinally and in the making of rope. The leaves have also been used for medicinal purposes as a diaphoretic and when young, are boiled and used in cases of syphilitic affectations (Drury, 1873), and in some cases are reported to be used to relieve nausea caused by overeating (Ochse, 1931). Young leaves are also eaten raw or steamed with rice. Medicinally, the fruit itself is considered to have excellent cooling properties. At maturity, the ripe fruits are very juicy and considered sweet but somewhat acid (Mowry and Toy, 1941). They may be eaten raw as a delicacy or made into jams, wines, sauces fbr fish and are often used in preserving (Brown, 1954; Burkill, 1935). Analyses of the fruit show that it is a good source of calcium and has a fair amount of iron (Maranon, 1935). In South Florida the fruit has enjoyed some popularity since 1939 and has been used there fbr the past fbur decades in pies, jellies, juices or eaten raw in a manner similar to that of raspberries, currants or the blueberries which it resembles (Fairchild, 1943). In 1972, at a luncheon for eight people, during which a pie made from the Antidesma berries was served, two persons at the table, after their first bite, complained that the pie was extremely bitter, hr; 1 Antidesma buniuslluunml a lrumnu hmmh h nlaln-lluuer ( N-mdll-tlmu-v Inn! 0 smtmnullrmt (Brown. 1954b 17 elk 1" ,. o ,0» 3%! if “ r. ‘. 'i 7 Il" I One of the most delicious and beautiful of the iellies for sale on the Miami market" Is made from the almost black fruits of this Antidesma bunius. When in fruit the tree is completely covered with these black clusters, making' It a spectacular sight. (Fairchild.1939) Fig. 2.--Tree of Antidesma bunius. 18 fm "’;'»;»~- "3 . 4 I I ~_‘.m‘— ' ' 5' “3“?) ' ~ » 1.. - » ‘— > I... )3; Tree of Antidesma bunius, on “The Kampong," that bears several bushels of fruit every August. It began bearing when six years old and might be compared with a giant currant bush for the clusters of fruit hang down in a similar way and make a delicious ielly that is compar- able in color and quality to current jelly. lt has several names in Java and the Philippines but its scientific name has become established here. Nathan Sands, who takes care of it, posing. (Fairchild.1939l Fig. 3.--Fruits of Antidesma bunius. 19 so much so that they considered it inedible. However, the other six persons at the table found the pie pleasant tasting, enjoyably edible and sweet. This incident, reminiscent in some manner of the divergent responses to PTC prompted a survey of taste responsiveness to this material by Henkin and Gillis (1977). In their study of 170 subjects, these workers not only fbund differences in taste perceptions to Antidesma fruit but also concluded that these differences appeared to be associated with the ability to taste PTC. In their study, responders to PTC and Antidesma extract were defined as those who described these solutions as bitter while nonresponders were defined as those who judged the solutions as either tasteless or of another taste quality (salty, sweet or sour). Subjects were also requested to record the intensity of their taste sensations on a scale of 1-100 based on their previous taste experiences. Of the 170 subjects studied, there were 115 PTC responders and 55 PTC nonresponders. Antidesma responders and nonresponders were 25 and 142 respectively. Among the 145 nonresponders to Antidesma, 67 judged the extract as slightly sour, 39 as sweet, 29 as salty and 10 could not designate any specific taste quality. A most interesting finding was the fact that among the 25 responders to Antidesma, there were 22 responders to PTC and among the 115 responders to PTC, there were no responders to Antidesma. Thus according to the data presented, three types of individuals were identified (Fig. 4): (l) PTC re5ponders-Antidesma nonresponders, (2) Antidesma responders-PTC non- responders, and (3) PTC nonresponders-Antidesma nonresponders. As can be seen, none of the individuals tested were responders (had bitter perceptions) to both Antidesma and PTC. These observations suggested 20 Key to Abbreviations: PTC-, Ad- PTC+ PTC Responders (30) PTC- PTC Nonresponders Ad+ Ad- Antidesma Responders Antidesma Nonresponders Fig. 4.--Antidesma and PTC Taste Responses (based on data from Henkin and Gillis, 1977. For details of data reported, see Appendix B). some type of interaction between the factors which determine the bitter response to PTC and those which are responsible for the bitter reSponse to Antidesma. Although these researchers inferred that this relation- ship may exist on a functional or genetic level, definitive conclusions regarding the nature of the interaction and inheritance pattern, if any, could not be formulated due to the relatively small numbers of indi- viduals sampled and the lack of appropriate family studies. In light of the above findings, the present study was proposed to: (1) sample larger numbers of individuals to establish frequencies for taste responsiveness to Antidesma by age, sex and racial groupings; (2) to confirm or refute the reported associations between Antidesma taste perceptions and taste responses to PTC and (3) conduct family studies to determine if the perceptions of Antidesma can be accounted for by a simple genetic hypothesis. The significance and utility of studies of this nature may be manyfold. If divergent responses to Antidesma are confirmed, this may stimulate a similar search and description of other naturally occurring substances for which such responses may be discovered and thus perhaps increase our understanding of the influences which these types of 21 substances exert on food and drink preferences and intake. If the Antidesma responses are fbund to conform to a specific genetic pattern, this may provide additional evidence that preferences for some food and drink may be determined, at least in part, by genetic factors. Further- more, confirmation of divergent Antidesma responses as reported earlier may lead to their use as markers descriptive of other human diversity parameters in population studies. Such markers may not only relate to food and drink preferences but also to drug responsiveness. Finally, as suggested earlier, such studies involving investigations of bitter responses may be useful in providing further information regarding the psychophysical and biochemical characteristics of bitterness, parti- cularly with respect to the number and nature of bitter receptors in humans. MATERIALS AND METHODS Sampling Procedures This study was conducted using random individual and family volunteers from which two major population samples were generated. Population I--Unrelated Subjects Individuals in this group consisted of volunteers from students and staff of Michigan State University. Staff members were sent memos or contacted directly to request their participation in the study. Student volunteers were primarily solicited from their Natural Science classes. Following a brief explanation of the purposes, risks and requirements of participation, individuals who agreed to volunteer were instructed to come in groups of two or three to a nearby sampling area fbr testing. Population II--Family Study Subjects Initially, one complete East Lansing subdivision consisting of 112 households was selected to approach for family volunteers. Letters were sent to these households in two stages. The first mailings, sent to approximately one-half of the households, introduced and explained the project (copy in Appendix C). Each of these households was subse- quently contacted directly at their home for fUrther explanations and to schedule them for testing if they agreed to participate. (Note: Care 22 23 was taken to make sure that sampling included only "intact" families, that is, those families in which both mother, father and their natural children were present in the household. Children under the age of seven were excluded to minimize the possibility of misclassification of taste perceptions and misinterpretation of instructions during sampling due to young ages.) The second mailings, sent to the other half of the subdivision households, contained similar information as the first mail- ing but also included more detailed explanations and a form for each family to complete and return (copy in Appendix C). Consenting families were contacted by phone for scheduling. Additional families were obtained by personal referrals from families who had already partici- pated in the study. TestinggProcedures Prior to any taste testing, an infbrmation and consent form was presented and thoroughly explained to all subjects (copy in Appendix D). Following the signing of the consent form, each individual was requested to complete the demographic portion of the survey questionnaire provided (copy in Appendix D). In cases where more than one subject was being tested concurrently, each person was then positioned so that they were unable to observe the other(s) and explanations of the testing pro- cedure were given. During this time, subjects were cautioned to refrain from making any verbal comments or gestures during the course of the taste sampling which might influence others being tested. Each subject was provided with unsalted crackers and a cup of distilled water to be used prior to the beginning of taste sampling and between each sample tasted to help neutralize taste flavors. 24 During the course of the taste sampling, subjects were seated and required to tilt their heads back and open the mouth with the tongue extended while keeping their eyes closed. Two to three drops of each solution to be tasted were flowed in turn over the surface of the tongue by means of glass droppers. Subjects were then instructed to taste the solution, record their perceptions by circling the appropriate taste quality (tasteless, salty, bitter, sweet or sour) for each solution and rate the intensity of their perceptions on a scale of 1—5 based on their previous taste experiences. If the solution was thought to be recog- nized by the subjects, they were requested to describe this in the appropriate place on the questionnaire fOrm. Taste Solutions Preparation and Processing The taste sampling panel was designed to assess taste percep- tions for eleven different solutions and included two samples of Anti- desma, three concentrations of PTC and six samples which served as con- trol solutions. Solution A--Antidesma I: An aqueous extract of Antidesma was prepared by gently pressing fresh berries of Antidesma bunius in fbur thicknesses of cheese cloth. The resultant liquid was filtered through #4 Whatman filter paper. The extract was stored at 0°C in 30 m1 aliquots and thawed when needed. Solution B--Sour Control: This solution consisted of commercially prepared natural strength reconstituted lemon juice (Realemon--Borden, Inc.) and was used at full strength. Bottles were purchased locally and stored at 4°C until used. 25 Solution C--Tasteless Control: Aliquots of distilled water were used for this solution. Solution D—-Salty Control: A l Molar salt solution was prepared by dissolving 58 grams of sodium chloride in one liter of distilled water. Solution E--Bitter Control: This solution consisted of 0.001 Molar quinine sulfate (Eli Lilly 8 Co.) and was prepared by dissolving 714.87 mg quinine sulfate in one liter of distilled water. Solution F--Sweet Control: A 0.5 Molar solution of sucrose was prepared by dissolving 171 gm of sucrose per liter of distilled water. Solution G--Antidesma II: Antidesma materials (skins, pulp, seeds, etc.) which remained from preparation of solution A (Antidesma I) were macerated by mortar and pestle to produce this solution. Aliquots derived were stored and used as indicated for solution A. Solutions H, I and J--Phenylthiocarbamide: A stock solution of PTC (Sigma Chemical) was prepared by dis- solving 81.25 mg per liter. This was used at full strength as solution J. Serial dilutions of the stock solution were made to give two addi- tional concentrations of 40.63 mg/liter (Solution 1) and 20.31 mg/liter (Solution H). Subjects were always required to taste the most dilute concentration first then progress up to the more concentrated solutions. 26 Solution K--"Fruit" Control: This solution consisted of commercially prepared unsweetened grape juice (Welch Foods, Inc.) and was used full strength from locally purchased bottles. As with the lemon juice in solution B, care was taken to open and use only small portions at a time to maintain fresh- ness . For taste sampling purposes, all solutions were stored and dis- pensed from 1 oz. dark-colored glass-dropper bottles and were stored at 4°C when not being used. Solutions were renewed every 3-4 days to insure freshness. RESULTS Overall Taste Perception Frequencies and General Demographic Data All sampling for this study was conducted between September 1979 and May 1980 and resulted in the testing of a total of 1,438 indi- viduals. Of this number 968 subjects (Population 1) represented unre- lated individuals and 470 subjects (Population II) were related. This latter group consisted of 112 two generation families, 3 three-generation families and 12 pairs of twins. During the course of the study, sampling was conducted at fre- quent intervals during the day. The time of testing depended on the time of availability of subjects and occurred between 8:08 a.m. and 11:30 p.m. The mean time of testing was 2:20 p.m. and the median was 1:30 p.m. Most of Population I (unrelated—students and staff) were tested during the weekday mornings and afternoons while most of Popula- tion 11 (families) were tested during the weekday evening hours as well as mornings and afternoons on Saturdays and Sundays. Ages of subjects ranged from seven to seventy-two years, with a mean age of 21.9 years (median = 18.2 years) and included 620 (43.1 percent) males and 818 (56.9 percent) females. Six different racial groups were also represented: 1,213 (84.4 percent) White/Caucasians, 198 (13.8 percent) Black/Afro Americans, 13 (0.9 percent) Chicano/ 27 28 Mexican Americans, 7 (0.5 percent) Asians, 6 (0.4 percent) Spanish American/Hispanics, and 1 (0.09 percent) American Indian. Other frequencies obtained included smoking status and elapsed time since last fbod eaten. The sample contained 258 (17.9 percent) smokers and 1,180 (82.1 percent) nonsmokers. Time of last food eaten by subjects prior to testing ranged from approximately 0.1 hours to 22.6 hours with one subject having not eaten in 50 hours. The mean elapsed time since last food eaten was 4.173 hours with the mode being 1.3 hours and the median 2.098 hours. The summary of taste responses obtained from the total popula- tion sampled is presented in Tables 1-4. Table 1a reports the taste perceptions of individuals for the control solutions, while the intensi- ties recorded fbr these items are shown in Table lb. As can be seen, for the fbur basic taste qualities (salty, bitter, sweet and sour), sweet and salty were most likely to be perceived as anticipated (98.6 percent and 97.5 percent). Expected perceptions of the tasteless con- trols were also at a high rate (98.1 percent). For the bitter control, 93.4 percent of subjects responded as expected. Of those misclassifying this control, a majority of these subjects judged it as sour (5 percent). Eleven subjects fbund the bitter control tasteless. The largest mis- classification occurred in perceptions of the sour control where 82.1 percent of subjects judged this as sour while 16.5 percent responded bitter. Perception of the "fruit control" shows that a majority of subjects judged this as sweet (73.2 percent) or sour (24.6 percent) while a few individuals fbund it bitter (2.0 percent) or salty (0.2 percent). No one found this solution tasteless. A comparison of inten- sities of controls as presented in Table 1b shows that the sweet control 259 ~m.n - gauge: ad.» . cameo: m.. . guano: ~¢.n . agave: ~c.o . sauna: mn.. . gauge: 4 - one: n . one: m . ova: c.e . one: o . 01o: m . ova: an.» . cue: n..n . can: oo.. . :aox _h.n . an»: nc.a . ego: a_.v . can: o.oo_ an.“ o.oc_ an." c.oo_ amen c.oc. anv— o.oo~ onq~ o.cc_ ”we. _auoe m.a~ am~ c.__ um” ~.om ”Na n.n~ he. o.o o a... See a «.nn co. ..o~ mac ~..~ hen a.~n one c.o o a.mn v_m v ~.a~ ~an o.~n as. m.n~ sac m.m~ can o.o o o.n~ cad n a.ma o- n.u~ nSN c.a cod o.°~ om“ m.o a 8.4 cc ~ m.o «a o.m cn_ m.v mo n.n av v.~ om m." - a o.o o mo.c a a.c a. ao.c a _.mm ~_v~ o.c o a 4 .oz 4 .oz a .82 a .02 a .o: « .oz :uOhucou aways: "ouueou «003m uouueou uouuwn ~ouueoo Aunum do." NCOU mnOMIuWIF nouueou nsom .m—ouueou mo noauumeouenuu.n~ oqnab -- a'.~ .o.o am.~ aa._ aa.a_ coauauacunna_uuax we xoeoacouu -ouo>o o.oo~ one" o.oo~ an.“ o.cc_ an4_ o.co_ anv~ c.oo. an." c.oc_ an." aqua» ~.o n _.c ~ n.o - n.5m ~ov_ ~.o n m.o a au_am o.~ a~ o.c a ..no nvn_ o.~ n~ o.c m m.o_ anN conga. ~.na ~mc~ o.om a_v_ o.o o n.o c m.o a a.o n_ noose o.c~ 4mm o.o a c.m Na o.c a o.o a ~.~a ~o_a “sow o.o o mo.o _ o.c __ mo.o _ a.mm _~4~ o.o o ano.ounae 5 .oz 3 .oz a .02 5 .oz a .oz » .oz :mouucou wanna: ~ouucou woorm nocueou nouumm ~ouucou >u_em nouueou mmo—oumab ~ouueou anew .naouueou we neeaunouuom ounahnt.n~ canah 30 was perceived as less intense than the other controls (mean intensity = 3.13) followed by the "fruit control" (mean intensity = 3.39) and the salty control (mean intensity 3.17). The greatest intensities were recorded for the sour control (mean intensity = 4.17) and for the bitter control (mean intensity = 4.06); median values and modes for intensities of controls are also recorded in Table la. (Note: Intensities of 0 represent individuals who judged controls as tasteless.) Perceptions of the two Antidesma preparations are recorded in Tables 2a and 2b. As can be seen, a majority of subjects judged Antidesma I (juice) as sweet (50.9 percent) or sour (36.0 percent) while 11.7 percent perceived this solution as bitter and a much smaller number found it tasteless (0.8 percent) or salty (0.6 percent). For Antidesma II (macerated material), most subjects perceived this as sour (45.9 percent), bitter (28.2 percent) or sweet (25.4 percent). Eight individ- uals judged this as salty while no subject found it tasteless. From these values, it will be noted that twice as many respondents reported Antidesma I as sweet as those fbr Antidesma 11 while nearly 2% times as many subjects judged Antidesma II as bitter as did those for Antidesma 1. Inspection of intensities reported for both antidesma solutions in Table 2b shows that individuals perceived Antidesma II as more intense (mean intensity = 3.1) than Antidesma I (mean intensity = 2.47). A comparison of taste responses to the Antidesma solutions and misclassification of control solutions was made to determine relation- ships between these two variables. For these analyses subjects were divided into three groups based on their perceptions of controls: (1) Individuals who made no errors (misclassifications), (2) Individuals who misclassified one control and (3) those who misperceived two or 31 Table 2a.--Taste Perceptions of Antidesma. Antidesma I Antidesma II No. % No. % Tasteless 12 0.8 0 0.0 Sour 517 36.0 660 45.9 Sweet 732 50.9 365 25.4 Bitter 168 11.7 405 28.2 Salty 9 0.6 8 0.5 Total 1438 100.0 1438 100.0 Bitter 168 11.7 405 28.2 Nonbitter 1270 88.3 1033 71.8 Table 2b.-~Intensities of Antidesma. Antidesma I Antidesma II No. % No. % 0 12 0.8 0 0.0 l 278 19.3 178 12.4 2 458 31.8 295 20.5 3 443 30.8 405 28.2 4 202 14.0 328 22.8 5 45 3.1 232 16.1 Total 1438 100.0 1438 100.0 Mean = 2.47 Mean = 3.1 Mode = 2 Mode = 3 Median = 2.44 Median = 3.11 32 more of the controls. Results of these comparisons are reported in Tables 3a for Antidesma I and 3b fbr Antidesma II. As seen in both tables, 75.9 percent of the total sample perceived all controls as expected (0 errors), 18.6 percent misclassified only one of the controls while 5.4 percent made two or more errors. Inspection of the row per- centages of each error category for each of the different taste percep- tions of the Antidesma solutions reveals that similar values were obtained. For example, if one considers the sweet responses to Antidesma I, 50.5 percent of individuals who made no errors, 51.5 percent who made one error and 55.1 percent of those who misclassified two or more controls judged this solution as sweet. Similar comparison of other perceptions for both Antidesma solutions produced similar results. Tests of association of misclassification of controls and perceptions of Antidesma results in the following values: For Antidesma I x Errors (Misclassifications of Controls) Cramer's V = 0.08247 Lambda (Asymmetric) = 0 with Error Dependent = 0 with Antidesma I dependent Lambda (Symmetric) = 0 For Antidesma II x Errors Cramer's V = 0.05469 Lambda (Asymmetric) = 0.00289 with Error Dependent = 0 with Antidesma II dependent Lambda (Symmetric) = 0.00089 (For explanation of rationale for use of these statistics, see Appendix.) Taste perceptions of the three PTC concentrations are reported in Table 4a. The greatest proportion of individuals judged each of these solutions as bitter (56.1-70.1 percent) or tasteless (24.2-39.2 percent) while a small number (4.7-5.7 percent) judged these as having other taste qualities (sour, salty or sweet). Application of the 33 Table 3a.--Comparison of Antidesma I Perceptions with Misclassifications (Errors) of Controls. *Count *Row % Antidesma I Perceptions Row *Column % Total *Total % Tasteless Sour Sweet Bitter Salty 8 382 551 146 5 0.7 35.0 50.5 13.4 0.5 1092 0 66.7 73.9 75.3 86.9 55.6 75.9 0.6 26.6 38.3 10.2 0.3 2 110 138 15 3 0.7 41.0 51.5 5.6 1.1 268 Errors 1 16.7 21.3 18.9 8.9 33.3 18.6 0.1 7.6 9.6 1.0 0.2 2 25 43 7 1 2.6 32.1 55.1 9.0 1.3 78 _>_2 16.7 4.8 5.9 4.2 11.1 5.4 0.1 1.7 3.0 0.5 0.1 Column 12 517 732 168 9 1438 Total 0.8 36.0 50.9 11.7 0.6 100.0 Cramer's V = 0.08247 Lambda (Asymmetric) 0 with Error dependent 0 with Antidesma I dependent 0 Lambda (Symmetric) *These designations apply to the four values (in the order tabulated) in each error category for each perception recorded in this table. These designations are also applicable to Tables 3b, Sa-c, 6b, 9b, 12b, 15b, 18b and 24a. 34 Table 3b.--Comparison of Antidesma II Perceptions with Misclassifica- tions (Errors) of Controls. Count Row % Antidesma II Perceptions Row Column % Total Total % Sour Sweet Bitter Salty 504 275 310 3 46.2 25.2 28.4 0.3 1092 0 76.4 75.3 76.5 37.5 75.9 35.0 19.1 21.6 0.2 125 70 69 4 46.6 26.1 25.7 1.5 268 Errors 1 18.9 19.2 17.0 50.0 18.6 8.7 4.9 4.8 0.3 31 20 26 1 39.7 25.6 33.3 1.3 78 3 2 4.7 5.5 6.4 12.5 5.4 2.2 1.4 1.8 0.1 Column 660 365 405 8 1438 Total 45.9 25.4 28.2 0.6 100.0 Cramer's V = 0.05469 Lambda (Asymmetric) Lambda (Symmetric) 0.00289 with Error dependent 0.00236 0.00089 with Antidesma II dependent 35 Table 4a.--Taste Perceptions of PTC. PTC (Low) PTC (Medium) PTC (High) No. No. No. % Tasteless 563 450 31.3 348 24.2 Sour 55 3.8 53 71 4.9 Sweet 2 5 .3 l 0.09 Bitter 806 920 1008 70.1 Salty 12 10 10 0.7 Taster Taster Taster 75.8 Nontaster 39.2 Bitter Nonbitter 43.9 Nontaster Bitter Nonbitter Nontaster 24.2 Bitter 70.1 Nonbitter 29.9 Table 4b.--Intensities of PTC. PTC (Low) PTC (Medium) PTC (High) No. % No. % No. % 0 563 450 31.3 348 24.2 1 184 119 8.3 88 .1 2 142 132 88 .1 3 184 146 122 .S 4 184 239 207 14.4 5 181 352 585 40.7 Mean = 1.85 Mean = Mean = 3.05 Mode = 0 Mode = Mode = 5 Median = 1.35 Median = Median = 3.85 36 traditional taster-nontaster classification produced an increased fre- quency of tasters and a corresponding decrease in nontasters with increasing concentrations of PTC. Based on the most frequently reported concentration of PTC (81.25 mg/l) employed to detenmine taster-nontaster status, the frequencies of tasters was 75.8 percent and nontasters was 24.2 percent. Classification by use of the bitter-nonbitter dichotomy reveals a similar increase in bitter responders and decrease in nonbitter responders with increasing concentration. Frequencies of those types of responders at the highest PTC concentrations (81.25 mg/l) results in frequencies of 70.1 percent bitter responders and 29.9 percent nonbitter responders. With respect to intensities recorded for the different PTC concentrations, Table 4b shows that the mean intensities increased with concentration from 1.85 for the lowest to 3.05 fbr the highest con- centration. The relationships of PTC perceptions for each concentration and misclassification of control solutions were determined and are reported in Tables 5a, 5b and Sc. Comparisons of the row percentages for each PTC perception show no significant differences between individuals who had no misclassifications and those who made one or more errors in per- ceptions of controls. Values fbr statistical tests of association (Cramer's V and Lambda) indicated no significant associations existed. Taste Perceptions and Age of Respondents Comparisons of age of respondents and taste perceptions to con- trols and experimentals were made. For purposes of these analyses, sub- jects were grouped in eight age categories: (1) 7-12 yrs, (2) 13-17 yrs, (3) 18-22 yrs, (4) 23-30 yrs, (5) 31-40 yrs, (6) 41-50 yrs, 37 Table 5a.--Comparison of Taste Perceptions of PTC (Low Concentration) with Misclassifications (Errors) of Controls. Count Row % PTC (Low Conc.-20.3l mg/l) Perceptions Row Column % Total Total % Tasteless Sour Sweet Bitter Salty 419 27 1 637 8 38.4 2.5 0.1 58.3 0.7 1092 0 74.4 49.1 50.0 79.0 66.7 75.9 29.1 1.9 0.1 44.3 0.6 106 17 1 141 3 39.6 6.3 0.4 52.6 1.1 268 Errors 1 18.8 30.9 50.0 17.5 25.0 18.6 7.4 1.2 0.1 9.8 0.2 38 11 0 28 1 48.7 14.1 0 35.9 1.3 78 3 2 6.7 20.0 o 3.5 8.3 5.4 2.6 0.8 0 1.9 0.1 Column 563 55 2 806 12 1438 Total 39.2 3.8 0.1 56.1 0.8 100.0 Cramer's V = 0.12138 Lambda (Asymmetric) Lambda (Symmetric) 0 with Error dependent 0.01582 with PTC Low Conc. dependent 0.01022 38 Table 5b.--Comparison of Taste Perceptions of PTC (Medium Concentration) with Misclassifications (Errors) of Controls. Count Row % PTC (Medium Conc.-40.63 mg/l) Perceptions Row Column % Total Total % Tasteless Sour Sweet Bitter Salty 338 27 2 721 4 31.0 2.5 0.2 66.0 0.4 1092 0 75.1 50.9 40.0 78.4 40.0 75.9 23.5 1.9 0.1 50.1 0.3 85 14 2 165 2 31.7 5.2 0.7 61.5 0.7 268 Errors 1 18.9 26.4 40.0 17.9 20.0 18.6 5.9 1.0 0.1 11.5 0.1 27 12 l 34 4 34.6 15.4 1.3 43.5 5.1 78 3.2 6.0 22.6 20.0 3.7 40.0 5.4 1.9 0.8 0.1 2.4 0.3 Column 450 53 5 920 10 1438 Total 31.3 3.7 0.3 64.0 0.7 100.0 Cramer's V = 0.15510 Lambda (Asymmetric) 0 with Error dependent 0 with PTC Med. Conc. dependent 0 Lambda (Symmetric) 39 Table Sc.--Comparison of Taste Perceptions of PTC (High Concentration) with Misclassifications (Errors) of Controls. Count Row % PTC (High Conc.-81.25 mg/l) Perceptions Row Column % Total Total % Tasteless Sour Sweet Bitter Salty 264 39 782 6 24.2 3.6 0.1 71.6 0.5 0 1092 75.9 54.9 100.0 77.6 60.0 75.9 18.4 2.7 0.1 54.4 0.4 65 14 186 3 268 Errors 1 24.3 5.2 69.4 1.1 18.6 18.7 19.7 18.5 30.0 4.5 1.0 12.9 0.2 19 18 40 1 .3 2 24.4 23.1 51.3 1.3 78 5.5 25.4 4.0 10.0 5.4 1.3 1.3 2.8 0.1 Column 348 71 1008 10 1438 Total 24.2 4.9 0.1 70.1 0.7 100.0 Cramer's V = 0.14706 Lambda (Asymmetric) Lambda (Symmetric) 0 with Error 0 with PTC High Conc. dependent 0 4O (7) 51-60 yrs, and (8) 61-72 yrs. Table 6a shows the distribution of "correct" and "incorrect" responses to the control solutions. As can be noted, the 18-22 years and 23-30 years age groups tended to misclassify all controls with greater frequency than other age categories. Elevated misclassification frequencies are also seen in the 7-12 years group fbr the sour and bitter controls. This trend is further suggested by data presented in Table 6b, which compares the misclassification of controls for the different age groups. As shown, the frequencies in the "no error" category for the age groups 18-22 years and 23-30 years are 71.1 percent and 60.0 percent respectively while in the 7-12 years group, 80.2 percent made no errors. These values may be contrasted with the percentages of individuals in other age groups who perceived the con- trols as expected which were 88.1-100 percent. Despite these apparent tendencies for certain age groups to misclassify the controls, statis- tical tests revealed no significant differences in perceptions of con- trols due to age (Cramer's V = 0.13904, Lambda = 0). Age related frequencies of taste perceptions of the Antidesma solutions are tabulated in Table 7a. For Antidesma I, a majority of subjects in all age groups perceived this as sweet (25.8-66.7 percent) or sour (22.2-52.6 percent). Individual percentages calculated for these perceptions for each group are not significantly different from the population average of 50.9 percent (for sweet) and 36.0 percent (for sour) nor do any apparent trends with age emerge. For bitter perceptions of Antidesma I the age groups of 7-12 years and 31-40 years had the highest frequencies of this response (20.6-20.8 percent), while fbr other age groups the percentage of bitter perception varied from 3.3- 17.8 percent. None of these values were found to be significantly 41 Aa-ev o o c.oo_ o o o o.oo_ a o o o.oo~ a o o c.oo~ a o o o.oc~ m Nn-—o .n Aa_.=v o o o.co— m~ m.o_ N m.oa «N o c o.co~ a“ o o c.oo~ G" c o o.co~ mu oo-_m .N ~c~_.=g o o c.oc~ a__ m.N n m.Nm m.~ N._ N n.om o__ o o o.oo~ c—— o.- n— a.mc mo~ cm-_v .o aamteu o.— ~ a.mm cm _.N N m.nm mm ~.N N m.Nm mm o o o.ooN No N.o o o.no _m cv-~n .m non-=3 n.n N N.om mN c.9N o c.oo vN 5.0 N n.nm oN n.n N N.om mN o.o~ o c.oe vN on-nN .v Aaomncu o.~ NN N.oa Nmm v.N NN o.Na can o.N oN ”.ma ova m.N vN o.Na «cm N._N o~N n.cN nmN NNIQN .n “dog-e. o c o.oo~ Non o.m m a.ma ca o.~ _ a.mm co~ o c o.oc— Non o.N n ~.Nm no NuanN .N Rom-c. o.~ N a.mm mo N.m m o.vo _o o.~ N c.om mo ~.N N a.Nm on o.v~ v— v.mu No Nuts .N a .02 N .02 o .02 a .02 I .02 a .02 o .oz o .02 a .02 o .02 uuouuooe— abouuou uuouuooeu uuouuoo uooauoueN uuoauoo uoouuooen uuouuou uoouuooe~ uuouuoo nhwuwonw< Nouueou uooam Nocueou eouuan nouueou xu_um Nouueoo «monounah Noaueoo anew .n.ouueou we eefiunouuom can: ueoveoanoa mo ou< we confines-outn.oo onauh 42 o a Aowuuossxmv owned; unopeonov ou< ea“: c a uneveoaov Nahum saw: o a Auwuuosexn 8.59lauu 8.88_ 8.8 8.. N.8 5.8 _.N 8.58 8.5 5.8 88888 888_ 8 8— 8.. 58 88 888 .8_ 88 .35888 8 8 N.8 8.8 N.8 8.8 N.8 N.8 8.8 I 8 8 8.N _.N 8.88 8.8 8.8 8.8 N x 85 8 8 8.8 8.N 8.8 N.N8 8.8 8.8 8 8 8 N 8 88 8 8 8 8.8 8.8 8.8 8.8 8.88 8.8 N.— 8.8— 8 8.88 8.8 N.5 8.88 8.NN 8.8 5.88 N 888888 88N 8 5.8 8.8 8.N 8.8 8.88 8.N 8.8 8 N 88 5 8 88N 5 88 8.8 N." N.5 8.8 8.. 8.58 8.8 8.8 8.85 8.88_ 8.88 8.88 5.88 8.88 8.85 8.88 N.88 8 N88— 8.8 8.8 8.8 8.8 8._ 8.88 8.8 8.5 8 5. 88_ 88 88 888 _8 55 .88 588 88-88 n58 88-88 A8. 88-88 “88 88-8N 588 NN-8N 588 58-88 5N8 N8-5 588 8 88888 838.8 a 8.5.8 no: aeoveoamom we ou< a sex 88:88 .8—ouueou mo Amuouuuu 8eowuauNuN888~u8qz can «coveonaox mo ou< mo eo»«uaauou-:.ao annoy .nvoocé I namuuimv cola.— uucoveonov : 3 8383 w~o_c.o .- ucovcomov one can: a a stain: ~35.— 3~o~—.o a > 81823.80 8: «Emu—88.882 a : _8< .5N88o.o a Au8uuollxm. auneag “unopeoaov 8 v< :88: 858°.o a Neoveonov one :88: o a Ao8euoasxn 8.803889 "— 8:8ov8ue< - 8 v< 113 8.8 8 8.8 8 N.8N 888 5.88 888 8.8N 888 8.88 N85 8.88 888 8.88 588 8 8 8.8 N8 888888 88.88 8 8 8 8 8.55 5 8.88 8 8 8 5.88 8 N.NN N N.NN N 8 8 8 8 N5-88 888 888.88 8 8 8 8 8.88 N 8.88 N 8.88 8 8.58 8 8.58 88 8.N8 8 8 8 8 8 88-88 858 8888.88 8.8 8 8 8 8.8N 8N 8.58 8N 8.88 88 8.88 88 8.88 55 8.88 88 8 8 8.N 8 88-88 888 858.88 8.8 8 8 8 8.88 88 8.8N 8N 8.88 88 8.8N 8N 8.88 88 8.N8 88 8 8 8.8 8 88-88 888 888.88 8 8 8 8 5.88 88 8.8 8 5.88 8 8.88 88 5.88 88 8.88 88 8 8 8.8 8 88-8N 888 8888.88 8.8 8 8.8 8 8.5N 58N 8.8 58 8.8N N8N 8.58 N88 8.N8 888 8.N8 888 8 8 5.8 5 NN-88 888 8888.88 8 8 8 8 5.5N 8N 8.88 88 8.8N 8N 8.88 N8 8.58 88 5.N8 88 8 8 8 8 58-88 8N8 888.88 8.N N 8.8 8 8.88 88 8.8N 8N 8.88 88 8.88 88 8.58 88 8.88 88 8 8 8 8 N8-5 888 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 88 88 8 88 88 88 8 88 88 88 8 8< 88 88 8 88 88 88 8 88 xuuam Nouuum acorn 830m nuouounah .ua8ov88e< we nebuunoouom £883 aeoveomnoz we ou< mo cenquenlou .q5 o—nah 44 different from the overall population frequency. The small numbers reported for the tasteless and salty perceptions were not conducive to analysis. With respect to age related frequencies of perceptions of Antidesma II individuals in all age groups most often judged this solution as sour (45.9%) or bitter (28.2%) with an appreciable nUmber (25.4 percent) reporting sweet perceptions. In all cases, with the exception of age groups 41-50 and 51-60, sweet perception frequencies were less than those of bitter. As with Antidesma I, no age trends are apparent for perception of Antidesma II nor are the frequencies of each specific perception reported for each age group significantly different from each other and from those found in the overall population. Despite the lack of age trends fer overall perceptions for Antidesma, a significant difference with age was found when the bitter- nonbitter classification for these solutions was employed. These data are reported in Table 7b. Analysis by X2 results in a probability of less than 0.05 for both Antidesma I and Antidesma II for age categories. Examination of X2 calculations however, reveal that for Antidesma I deviations of the age groups 1 (7-12 yrs), 3 (18-22 yrs), 5 (31-40 yrs) and 6 (41-50 yrs) made the greatest contributions to the X2 value while for Antidesma II, greatest deviations from expected were found for the age groups 4 (23-30 yrs) and 8 (61-72 yrs). Such results again fail to substantiate definitive age trends. Age group categories were compared with respect to their per- ceptions to the three concentrations of PTC and are presented in Table 8a. For the lowest concentrations of PTC, it will be noted that the overall average frequency for the tasteless perception was 39.2 45 Table 7b.—-Comparison of Age of Respondent with Antidesma Bitter- Nonbitter Perceptions. Antidesma I Antidesma II Age Groups (Years) Bitter Nonbitter Bitter Nonbitter No. % No. % No. % No. % (1) 7-12 20 20.8 76 79.2 33 34.4 63 65.6 (n=96) (2) 13-17 16 15.8 85 84.2 28 27.7 73 72.3 (n=101) (3) 18-22 87 9.0 881 91.0 267 27.6 701 72.4 (n=968) (4) 23-30 1 3.3 29 96.7 14 46.7 16 53.3 (n=30) (5) 31—40 20 20.6 77 79.4 30 30.9 67 69.1 (n=97) (6) 41-50 21 17.8 97 82.2 24 20.3 94 79.7 (n=118) (7) 51-60 2 10.5 17 89.5 2 10.5 17 89.5 (n=19) (8) 61-72 1 11.1 8 88.9 7 77.8 2 22.5 (n=9) Totals 168 11.7 1270 88.3 405 28.2 1033 71.8 2 2 ._ X7 - 30.29 X7 - 25.25 p < 0.05 p < 0.05 .88888.8 . 88888822288 888.88 8888888288 828828 882 288- 8 . 888888288 888 288: 88N88.8 . 888888222888 8882-8 858888.8 . 8 8.88-888 8888- 8N.88 . 2882 882 .8 . 88888822288 888.88 8888888288 88828 882 288- 8 . 888888288 .8. 2888 8 . 888888.22888 888-88 888888.8 . 8 ..88-888 8888. 88.88 . 882 882 46 .88888.8 . 888888-2288 888288 8888828288 88888 882 8888 88888.8 . 888888288 88. 2882 8 . 888888222888 888288 888888.8 . 8 8.88-888 8888- 88.8N 8 :88 882 5.8 88 5.8 88 8.8 N8 8.85 8888 8.88 8N8 8.88 888 88.8 8 8.8 8 8.8 N 8.8 85 5.8 88 8.8 88 N.8N 888 8.88 888 N.88 888 888888 88.28 8.88 8 8 8 8 8 8.88 8 8.88 8 8.88 8 8 8 8 8 8 8 8.88 8 8.88 8 8.88 8 8.88 8 N.NN N 8.88 8 N5-88 .8 888.88 8 8 8 8 8 8 8.85 88 8.85 88 5.85 88 8 8 8 8 8 8 8.88 N 8.88 N 8.88 8 8.88 N 8.88 N 8.88 N 88.88 .5 8888.88 8.8 8 8 8 5.8 N 8.85 88 8.88 88 8.58 88 8 8 8 8 8 8 8.8 5 8.8 8 5.8 N 8.NN 5N 8.8N 88 8.88 88 88-88 .8 858-28 8 8 8 8 8 8 8.85 88 8.88 88 8.N8 88 8 8 8 8 8 8 8.N N 8.N N 8.8 8 8.8N 8N 8.88 N8 8.88 88 88-88 .8 888-2. 8 8 8 8 8.8 8 5.88 8N 8.88 88 8.88 88 8 8 8 8 8 8 5.8 N 8 8 8.8 8 5.8N 8 8.88 N. 8.88 88 88-8N .8 8888.28 8.8 8 8.8 8 8.8 8 8.85 588 8.88 888 5.88 888 88.8 8 8.8 8 N.8 N N.8 88 8.8 88 8.8 8N 8.8N 88N 8.N8 888 8.88 888 NN-88 .8 . . 8888.88 8 8 8 8 N N 8.N N 8.85 85 8.88 88 8.88 88 8 8 8 8 8 8 8.8 8 8.8 8 8.8 5 8.88 8N 5.8N 8N 5.88 N8 58.88 .N . . 888828 N 8 8 N 8 8 8.N N 8.88 88 8.88 88 8.58 88 8 8 8.8 8 8 8 8.88 88 8.88 88 8.88 88 8.5N 8N N.88 8N 8.88 58 N8.5 .8 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 8 .82 .88 82 8.2 .58 .8882 8.82 :3 .88 82 882 :3 .88 82 882 :3 88882 882 8.3 8888828 882 882 882 882 882 882 882 882 882 882 882 882 882 882 882 822888 888 8883 .8388- uooxm 28.2.8 88.83828. .982 no 2088108802 ‘88: 83089958.. 80 ou< no 29888.8-50--.o- .8828 47 percent and that the frequency of nontasters varied from 10.5 percent (ages 51-60) to 43.3 percent (ages 23-30). These two extreme frequencies in the tasteless category represented only a small number of subjects however. Frequencies fer the bitter perception ranged from 33.3 percent (three of the nine subjects in age group 61-72 years) to 62.9 percent (ages 31—40 years) with an average bitter frequency of 56.1 percent. While the overall frequency of sour perception was 3.8 percent, age groups which showed the greatest tendency to judge this PTC concentra- tion as sour were groups 1 (7-12 yrs), 7 (51-60 yrs) and 8 (61-72 yrs) reporting frequencies of sour greater than 10 percent. Both individ- uals who perceived this as sweet were in age group 3 (18-22 years). Additionally, there were 12 subjects (0.8 percent) who judged this solu- tion as salty. Cramer's V and Lambda statistical tests of association however, revealed no significant differences for age groups for this lowest PTC concentration. For the medium concentration of PTC, the frequencies of percep- tions of tasteless varied from 10.5 percent (ages 51-60) to 40.0 percent (ages 23-30) with an overall frequency of 31.3 percent for this percep- tion. 0f the five individuals who judged this solution as sweet, four were in age group 3 (18-22 yrs) and one in group 1 (7-12 yrs). The average frequency of the sour response was 3.7 percent with age groups 1, 7 and 8 reporting a sour frequency greater than 10 percent. The percentages of the bitter perception varied from 44.4-78.9 percent while the average frequency was 64.0 percent and ten individuals (0.7 percent) recorded salty perceptions. As reported fOr the previous PTC concentra- tion, calculations of statistical tests for the medium concentration of PTC show no significant associations with age. 48 Table 8a also shows age related perceptions reported for the high concentration of PTC used. As reported previously, the average frequency of nontasters (tasteless) decreased to 24.2 percent but ranged from 10.5 percent for age group 7 (51-60 yrs) to 33.3 percent for group 8 (61-72 yrs). While the average frequency of sour responders for this solution was 4.9 percent, age groups which reported a sour frequency of greater than 10 percent were groups 1 (7-12 yrs), 7 (51-60 yrs) and 8 (61-72 yrs). Only one individual judged this solution as sweet (age group 3) and ten subjects reported salty perceptions. Frequencies of bitter responders varied from 44.4 percent fer age group 8 to 78.9 per- cent fbr age group 7 with an overall average frequency of 70.1 percent. No significant associations were fbund for the different perceptions of the PTC high concentration with age. To facilitate subsequent correlations, age related PTC percep- tions were compared with respect to taster-nontaster and bitter-nonbitter status and are reported in Table 8b. For the low concentrations of PTC taster frequencies ranged from 44.4 percent for age group 8 (61-72 yrs) to 89.5 percent for age group 7 (51-60 yrs) with an average overall frequency of tasters of 60.8 percent. For the medium PTC concentration, minimum (60.0 percent) and maximum (89.5 percent) taster frequencies were obtained fbr age group 4 (23-30 yrs) and 7 (51-60 yrs) and an average frequency of 68.7 percent. Corresponding minimum (66.7 percent) and maximum (89.5 percent) taster values fbr the high concentration of PTC were fbund in age group 8 (61-72 yrs) and 7 (51-60 yrs) and the average taster frequency was 75.8 percent. Age differences with respect to PTC taster-nontaster status were not statistically signifi- cant (p > 0.05). 49 .00.0 A 0 .50.05 . 5x emcee A n QN—ov ” 5x ”555- 05.50 . 005: 050 5 5 .00.0 A 0 .00.05 - M5 .00.0 A 0 .05.0 - M5 5550- 00.00 . 00: 000 .00.0 A 0 .00.0 . M5 .00.0 A 0 .00.55 a M0 "550- 50.05 . :00 050 0.05 000 0.00 050 0.00 500 5.05 0005 0.00 050 5.00 000 5.05 000 0.50 000 5.00 000 0.05 0005 5.00 000 0.00 050 05.000 Aoucy 0.00 0 0.00 0 5.00 0 0.00 0 0.00 0 0.00 0 0.00 0 5.55 5 0.00 0 5.00 0 0.55 5 0.00 0 55-50 .0 nm—acv 5.55 0 5.55 0 0.05 0 0.05 05 0.05 05 5.05 05 0.05 5 0.05 5 0.05 5 0.00 55 0.00 55 0.00 55 00-50 .5 5055.00 5.05 00 0.50 50 0.50 00 0.05 00 0.00 50 0.50 00 0.55 55 0.05 00 0.00 00 5.55 50 0.55 00 0.50 55 00-50 .0 550-00 0.05 05 5.00 00 5.50 00 5.55 00 0.00 00 0.50 50 0.05 05 0.00 50 5.00 00 5.05 55 0.50 00 0.00 50 00-50 .0 500.0. 0.00 05 0.00 55 0.00 05 5.00 05 0.00 05 0.00 05 5.05 0 0.00 55 0.00 05 0.05 55 0.00 05 5.00 55 00-05 .0 5000-05 0.05 505 5.00 000 0.00 050 0.55 500 0.00 050 5.00 000 0.05 005 5.50 550 0.00 000 0.05 505 0.50 500 0.00 050 55-05 .0 5505.0. 5.05 55 5.50 50 0.00 50 0.05 05 0.00 00 0.00 00 0.05 05 5.05 05 5.50 50 5.00 50 0.55 55 5.00 00 55-05 .5 “om-cu 5.50 00 0.00 00 5.50 00 0.00 00 5.00 50 0.50 00 5.55 05 5.00 05 0.00 50 0.55 05 0.00 50 0.50 00 .005 55-5 .5 0 .02 0 .02 0 .02 0 .02 0 .02 0 .oz 0 .02 0 .02 0 .02 0 .02 0 .02 0 .oz 005: 00: :00 :05: 00: :00 :05: 0o: .05 :05: 00: :00 505.050 0.5... 0.: 0.5.5 05.. 0.: 0.: 0.: 0.: 05.5 0.: 0.: 0.5.5 05500.6 0055 heuu5acoz nouu5n «poem-ace: abounnh .0:05uaouuoa nouuuncoz-uouu5n at. hounuucoz-uounnh uh; Aug: acovconnoa uo ou< mo conauunloo-u.an o—nnb 50 Table 8b also shows that fbr the three concentrations of PTC, low, medium and high, the average bitter responders were 56.1 percent, 64.0 percent and 70.1 percent respectively. Comparisons of PTC percep- tions with age shows that fbr each concentration of PTC the youngest subjects (group 1, 7-12 yrs) and oldest subjects (group 8, 61-72 yrs) were fbund to have the lowest frequencies of bitter responders. Corresponding frequencies for other age groups were varied with no dis- cernible age trends. Indeed, X2 analysis shows that with respect to age the bitter-nonbitter PTC status for the concentrations used was not significant (p > 0.05). Taste Perceptions and Sex of Respondents As reported previously, taste perceptions of 620 males and 818 females were assessed. Perceptions of the control solutions by sex are reported in Table 9a. In all cases except for the sweet control, females were less likely to report "incorrect" (misclassifications) perceptions when compared to males. The greatest differences of incor- rect perceptions between the sexes appears in the misclassification of the bitter control in which the "error" rate fer males (9.5 percent) is over two times that of females (4.4 percent). For the sour con- trol, where overall misclassifications were more frequent, males were about 80 percent more likely to incorrectly perceive this control (Error rate was 20.2 percent fbr males and 16.1 percent for females). In spite of these apparent male-female differences in perceptions of the controls, statistical tests of associations for error categories (0, 1 and 2) as reported in Table 9b, shows no significant associations of misperceptions with sex of respondents. 51 Table 9a.--Comparison of Sex of Respondent and Perception of Controls. Males (n=620) Females (n=818) No. % No. % Sour Control Correct 495 79.8 686 83.9 Incorrect 125 20.2 132 16.1 (Overall frequency of Sour Control misclassification = 17.9%) Tasteless Control Correct 604 97.4 807 98.7 Incorrect 16 2.6 11 1.3 (Overall frequency of Tasteless Control misclassification = 1.9%) Salty Control Correct 604 97.4 798 97.6 Incorrect 16 2.6 20 2.4 (Overall frequency of Salty Control misclassification = 2.5%) Bitter Control Correct 561 90.5 782 95.6 Incorrect 59 9.5 36 4.4 (Overall frequency of Bitter Control misclassification = 6.6%) Sweet Control Correct 613 98.9 805 98.4 Incorrect 7 1.1 13 1.6 (Overall frequency of Sweet Control misclassification = 1.4%) 52 Table 9b.-~Comparison of Sex of Respondent with Misclassification of Controls. Count Row % Sex of Respondent Row Column % Total Total % Males Females 454 638 41.6 58.4 1092 0 73.2 78.0 75.9 31.6 44.4 118 150 Errors 1 44.0 56.0 268 19.0 18.3 18.6 8.2 10.4 48 30 ‘3 2 61.5 38.5 78 7.7 3.7 5.4 3.3 2.1 Column 620 818 1438 Total 43.1 56.9 100.0 Cramer's V = 0.09113 Lambda (Asymmetric) Lambda (Symmetric) O with Error dependent 0.02903 with Sex dependent 0.01863 53 Sex differences with respect to perceptions of Antidesma are tabulated in Table 10a. For Antidesma I, proportions of males and females in each taste perception category were quite similar in that the differences between the sexes ranged from 0-2.l percent. For Antidesma II, although a wider range of taste perception differences for males and females was fOund (O-S.0 percent) such differences were not striking. Thus fOr both Antidesma I and Antidesma 11, overall fre- quencies fbr each perception were not significantly different when male-female comparisons were made (p > 0.05). Further analysis of sex differences for Antidesma perceptions as reported in Table 10b produced dissimilar results. As can be observed, comparison of sex of respondent with respect to bitter and nonbitter perceptions revealed that for Antidesma I, no significant sex differences were obtained (xi = 1.519, p > 0.05). For Antidesma 11 however, bitter-nonbitter perceptions of males and females are statis- tically significant (xi = 4.315, p < 0.05). Frequencies of taste responses fer each of the three concentra- tions of PTC by sex are reported in Table 11a. These data show that for each PTC concentration, a greater proportion of’males found these solutions tasteless, sour or salty when compared to females. Conversely, females were more likely to report bitter or sweet perceptions than males. The overall perceptions fOr the different sexes were not signifi- cant (see Cramer's V and Lambda values). Results compiled in Table 11b show the proportions of males and females who were classified as PTC tasters or nontasters as well as those who were bitter and nonbitter responders. It is apparent from these data, that for all PTC concentrations, a greater proportion of 54 00.0 v 0 .500.0 a 55 00.0 A 0 .050.5 . 5 x 5 5 c.5n nmo5 ~.m~ mcv n.w¢ o-5 0.55 mo5 5auoh 5050-05 5.00 050 m.cn 005 0.55 555 0.55 005 0050900 5050-05 5.05 now 0.05 505 0.0» mmm m.o5 me 0050: 5 .oz 5 .oz 5 .62 a .62 5355055562 .5055: 505535502 .5055; 55 .20 0505552 5 div-0505552 .msuaum nouu5ncoz-uouu5m nanoe5uc< :55: unovcodnoz we now mo 00055.5360 .ao5 o5pnh mo.o x d .v5~.m - Mx u55 v< mo.c A a .005.~ - “x .5 1< 5.c o o.m 5.5 0.0 5.5 ~.5 n.c o v.o a :5 «00:0505555 5n5nn:5 m.c c 0.9 m 0.00 505 0.55 005 5.05 005 o.om gov v.00 550 «.00 005 o o o.5 a 005alun 5050.05 o.o v o.o v 0.05 505 0.95 no v.5~ o55 5.5m 050 o.ov 055 5.00 v- o o o.c v 0054: a .02 a .62 a .62 5 .02 a .62 5 \..oz 5 .oz 5 .62 5 .62 o .oz 55 v< 5 v< 55 v< 5 u< 55 v< 5 v< 55 v< 5 v< 55 v< 5 u< Au5um 50555: uoozm 530m 00050503h .olnov5u:< «6 00055500505 :55: «coveomnoz uo xom uo :ou5uamnou .uc5 o5nah 55 =¢c~o.c I 555.000.... .5025 .50.. 55.. 55500.0 . 555.5005. .3. 55.5.. 50050.0 - 505558.125 02-15 05000.0 . 55 0.55.6 ”:5... 0:55.55 0 5005.02.05. 5.8. u: 0.55.. o .. 5550505500505. ~00 £55.. 5558.: 0 50555125 oval-.5 55000.0 . 0 0.50.0.6 5:50- 20506 a 55.05.00.505. .65 p: 55.. 0 . 5:055:09! :0 5:5- 5:36 .- 5o5ualn<5 0303.5 00005.0 _. > 0.53.20 H:5... Roda «Emu 03.1.— 5052105 03.35 595.313 1.x...— 05.500 .505: or. no.9: to: E 5nd: )3 P: 050.5 0.0 0 0.0 0 0.0 0 0.55 5550 0.00 000 0.00 500 00.0 5 0.0 0 5.0 5 0.0 50 0.0 05 05 05 5.05 005 5.00 005 0.50 500 005...: 5050.5 0.5 0 0.5 0 0.5 0 0.50 550 0.50 30 0.50 050 0 0 5.0 5 0 0 0.0 00 0.0 05 0.0 50 0.05 005 0.50 005 0.50 005 005! o .35 o o o 6.. o .35 o .92 o .02 o .62 o .02 o .0: a .0: o .95 o .02 o .95 o .95 .55 55. 00... .55 .55. :5 5.85 :3 .555 :5 5;: .65 .5. 5: 5:: :3 .50 :5 5...... .65 u: .0: 0: pr. or. at E0 555.. or. pr. or. of. 0.5.5 0.5.5 95... 55.5 .5355. 59.5.6. .5655 0005.050: .9: 5o 0.855.535..— a... 5.5-55.3.500- uo no» no .30 5500-8- - .155 0555-... 56 00.0 A 0 .000.5 . M05 00.0 x 0 .5050 . M5 .. 5550.. 05.505 550555 00.5 00.0 x .5 .550.5 .. wx 00.0 0 0 .050.5 .. M5 - 5550.. 00.005 5.0: 0.... 00.0 v 0 .000.0 0 M05 00.0 x 0 .000.5 0 M5 - 5550.. 50.055 8.5 0.: 0.05 000 0.00 050 0.00 500 0.00 050 5.00 000 5.05 000 0.50 000 5.00 000 0.05 0005 5.00 000 0.00 050 5000.— 5050.05 5.05 505 0.00 505 5.50 000 0.00 000 0.00 500 5.05 005 5.00 005 0.50 500 0.05 050 0.00 550 0.50 550 0.50-0.5 5050.5 5.50 005 5.00 005 5.50 005 0.50 000 0.50 050 0.05 005 0.50 005 0.50 005 5.05 000 5.50 050 5.00 000 0050.5 ' a; ’ so: ' to: ' .02 ' ooz * I; ' Cg ' .2 ' no: ' no: i no: 550555 5.0: :3 5.05.. :3 550055 00: :3 550555 5.00. .65 0.00 0.: 0.5.5 0.: 00.0 0.555 p: or. p: 00.5 2.5 nouu5acoz 50555: nuounnueoz uuounah .msu-um 505555002-505555 can noun-acoz-uounnb up; can acov=050o¢ mo now mo con5u¢nlou:-.n55 o5nah 57 females was found to be tasters and more often perceived each PTC solu- tion as bitter when compared to males. No significant differences were found between the sexes however, for any of the PTC concentrations when the taster-nontaster classification was employed. A similar lack of significant sex differences was found fbr bitter-nonbitter responders for the medium and high PTC concentrations. However, these responses to the low concentration of PTC did result in significant differences between sexes (x: = 6.364, p < 0.05). Taste Perceptions and Race of Respondents Each subject participating in this study assigned themselves to one of six racial/ethnic group categories (White/Caucasian, Black/Afro Americans, Chicano/Mexican American, Spanish American/Hispanic, American Indian or Asian/Pacific Islander). Comparison of perceptions of con- trols by racial groupings are tabulated in Table 12a. Inspection of these data shows little overall racial differences in expected percep- tions of these solutions. It will be also noted that in instances of apparent striking racial differences from the average "correct-incorrect" frequencies (e.g., Spanish American/Hispanic perceptions of the bitter control) small numbers of individuals in these groups appear to be responsible for the deviations. Further substantiation of lack of racial differences in perceptions of controls is shown by data reported in Table 12b in which misclassifications of controls are compared by race fer each error category. As seen, racial frequencies for these error categories are comparable except in cases where racial groupings con- tained small numbers of subjects. Statistical tests of association 58 0.5 05 0.00 0505 0.0 00 0.00 0005 0.5 00 0.50 5005 0.5 55 5.00 5505 0.55 505 5.50 5055 0.0005 55-05 0.05 5 5.00 0 0 0 0.005 5 0 0 0.005 5 0 0 0.005 5 0.05 5 5.00 0 0050< .0 55.05 0 0 0.005 5 0 0 0.005 5 0 0 0.005 5 0 0 0 005 5 0 0 0.005 5 005005 .0< .0 50-05 . . . . . . . 05000052 0 0 0 005 0 0 00 5 5 00 0 0 0 0 005 0 0 0 0 005 0 5 05 5 0 00 0 5.00 2050000 .0 505-05 . . . . . . . .00 000500: 0 0 0 005 05 5 5 5 0 50 55 0 0 5 0 00 55 0 0 0 005 05 0 0 0 005 05 50000500 .0 5005.05 Q 0 I I O O o C O O 0: Ohu< 0 5 5 0 00 005 5 0 05 0 50 505 0 0 0 0 50 505 0 5 0 0 00 005 5 05 05 0 00 055 550050 .5 50555.05 0.5 55 0.00 0055 0.0 05 5.00 5055 0.5 05 5.50 0055 0.5 05 5.00 0055 5.05 555 0.50 000 0050Mwmuw .5 0 .02 0 .02 0 .02 5 .02 0 .02 0 .02 0 .02 0 .02 0 .02 0 .02 nacho 0.5.500 uu 5.5— uuohhoo uuotOUc— 000.5500 9005.505555— uUOhLOU Huck-5005: you-5.500 0005:5005: “UP—.500 Uwczum\vuu¢ 5 0.5 acou 5095 5055060 565555 5055:6u >550m 5ouucou 0005oumah 5ohucou 550w .05ouucou uo mco5unoocoa can ucovconmoa mo coax uo :o055-nloo-.n~5 05559 59 o u namuuueexmv «vs-Ha acovconuu ovum :55: o n acovcoaov nouum :55: c u Aufiuuolzxm m.h0£auu 0.005 0.0 5.0 0.0 0.0 0.55 0.00 50505 05.05 5 5 0 5.5 005 255 5558 0 0 0 0 0.0 0.0 0.0 n 0 0 0 0 0.0 5.0 5 A 05 0 0 0 0 0.55 0.00 0 0 0 0 0 00 5.0 0 5.0 5.0 0.5 0.05 0.05 0.05 0 0.00 5.55 5.55 0.05 5 050500 005 5.0 0 5.5 5.5 5.55 0.00 5 0 n n 00 055 5.0 5.0 5.0 5.0 5.05 0.50 0.05 0.55 0.005 0.00 0.05 0.55 0.05 0 5005 0.0 5.0 0.0 0.0 5.05 5.00 m 5 n 05 005 050 50000505 .000505 005005505 05000052505 .-< 0005502505 0005500< 005<555 005000000555 5 50505 5305 5:052 5055...! 5.5 5505500 50085550 5050050 535552 0 5358 00¢ 0 30¢ acovconmoz mo coax ucsou .n5ouucou mo Anuouumu :o5uu05uauma5unmz and «covconnou mo ouuu uo canuuonlou--.n~5 o—auh 60 confirmed lack of significant association of race and errors for per- ceptions of control. Table 13a reports the perceptions of the Antidesma solutions by race. For Antidesma I, where perceptions of tasteless and salty were the lowest responses reported, only whites and blacks are represented. In general, a greater proportion of individuals of all races (except Chicano) judged this solution as sweet with an average frequency of 50.9 percent. Racial frequencies for the sour response ranged from 26.8 percent fOr Asians to 46.2 percent for Chicanos. (Note: Sample size fbr these groups are small.) For the bitter perceptions of Antidesma I, none of the 7 Spanish Americans, 6 Asians nor the single American Indian reported this response. This may be contrasted with the 3 Chicanos (23.1 percent), 53 Blacks (26.8 percent) and 112 whites (9.2 percent) who reported bitter perceptions. Cramer's V and Lambda statistics indi- cate no overall significant associations of Antidesma I perceptions with race. With respect to race related perceptions for Antidesma II, Table 133 shows that most often racial groups judged this solution as sour (average frequency was 45.9 percent). Exceptions to this generalization occurred for Blacks and the single American Indian sub- ject. As can be seen, over one half (50.5 percent) of Blacks and the American Indian perceived Antidesma II as bitter as compared to bitter frequencies of 14.3-24.5 percent fbr the other racial groups. Sweet perceptions for whites and blacks were 27.4 percent and 13.1 percent respectively while this perception reported by other racial groups yielded 0-50 percent due to small numbers of subjects. As with Anti- desma I no significant overall associations of race and perceptions of Antidesma II were fbund. (51 om5mo.c u -mcc.o I 5u5cuoesxm5 cabana 520000500 55 u< :55: 05500.0 0 520000500 0000 :55: 0 n 5u5uuoauxm<5 unneu5 .50055.0 0 > n.5ueacu 555 0500055:< u 55 u< 505305.58 05.55.00,. 500000000 5 00 :55: 50500.0 0 500000500 0002 :55: 00000.0 . 5055502250<5 000005 .00555.0 u > 0.502050 55 020005500 0 5 00 0.6 a 0.0 o ~.0N mo. 5.55 $05 v.m~ men m.om Nnn a.mv coo c.0m 55m c o m.o N5 m5uuoh 55005 c c c c n.v5 5 o o m.~v n 9.55 m m.~v n o.w~ N o o o c :u5m< .o 55005 c o o o 0.695 5 o c o c c.oo5 5 o o o o o o o o :Q5vc5 .l< .m 500:5 . . . . u5=aam5= 0 0 0 0 5 05 5 0 0 0 00 n 0 005 0 n 55 5 0 0 0 0 0 0 5.00 2050000 .0 55.05 . . . . . . .£< :uo5xo: 0 0 0 0 m 05 m 5 05 n 5 5 5 0 05 0 0 mm 5 5 00 0 0 0 0 0 50000520 .5 Aao5lcu . . . . . . . . . .l< ouu< o 5 N m N m m cm 005 m ON mm 5 n5 ow v we we 0 mm as w on mm c o m 5 n \xua5n .N An5~5ncv . . . cu5maozau m c o n o v m.v~ th ~.a ~55 v.n~ Nmn 5.~m Nno n he mum 0.0m cmv o o 5.0 m \Ou5 . .52 5 a .02 a .02 p .02 o .02 a .02 a .02 a .02 a .02 .02 a .02 nacho 55 u< 5 u< 55 u< 5 n< 55 u< 5 v< 55 u< 5 ad 55 n< 5 v< 05::um\uuu¢ >u5am 505555 uoorm 530m nmo5oumoh .aamuv5uc< mo n=o5unoouom :55: acovcoanou we con: 50 can5uualouuu.qn5 o5pah 62 Because of the relatively small sample sizes of several racial categories, subsequent analyses of taste perceptions of Antidesma and PTC by race were restricted to the white and black racial groups. (General frequency data however is presented for all racial groups sampled.) Table 13b presents racial perceptions of Antidesma I and II when classified as bitter and nonbitter. As shown, the bitter percep- tion of Antidesma I fbr Blacks was nearly three times that of Whites (26.8 percent versus 9.2 percent) and fOr Antidesma II, Blacks were more than twice as likely to perceive this solution as bitter when com- pared to Whites (50.5 percent versus 24.5 percent). These differences are shown to be highly significant by Chi-square analysis (p < 0.05). Overall racial perceptions of the three PTC concentrations are reported in Table 14a. Based on the indicated statistical treatments, no significant associations with race were established for any of the concentrations of PTC. It will be further noted however, that for the major racial categories represented, for each concentration Blacks were less likely to judge PTC as tasteless but more likely to perceive these as sour or bitter when compared to Whites. Sweet and salty responses fOr these groups were similar although small numbers of subjects record- ing these perceptions prohibited precise generalizations. Comparisons of racial perceptions of PTC with respect to the taster-nontaster and bitter-nonbitter dichotomies are presented in Table 14b. Taster frequencies ranged from 57.1-100 percent fbr all PTC concentrations (extreme values were obtained from the smaller racial groups). When White-Black taster frequencies are compared, the differ- ences between these groups increase with concentration. Thus, at the low concentration, values obtained were 60.6 percent for Whites and 62.1 63 Table l3b.—-Comparison of Race of Respondent with Bitter-Nonbitter Antidesma Perceptions. Antidesma I Antidesma II Race/Ethnic Bitter Nonbitter Bitter Nonbitter Group ‘________ __________ __________ No. % No. % No. % No. % White/Caucasian 112 9.2 1101 90.8 297 24.5 916 75.5 (n=1213) Black/ 53 26.8 145 73.2 100 50.5 98 49.5 Afro-American (n=198) Chicano/ 3 23.1 10 76.9 5 38.5 8 61.5 Mexican Am. (n=13) Spanish Am./ 0 0 6 100.0 1 16.7 S 83.3 Hispanic (n=6) Am. Indian 0 O 1 100.0 1 100.0 0 0 (n=1) Asian 0 0 7 100.0 1 14.3 6 85.7 (n=7) Total 168 11.6 1270 88.4 405 28.2 1033 71.8 Analysis¢of White/Caucasian and Black/Afro-Americans Antidesma responses Antidesma I: xi 50.63, p < 0.05 Antidesma II: xi 56.97, p < 0.05 64 4.5.58.0 0 505.5513 .3 55550352505 55:: PE .555) nn~86 I .555053525050 00.: £55) a - 595551-45 oi 55.85.55 u > 2.5-I959 ”2" «~53 55.5: D: .2050... . 505053-505 03.3 555002500 5.02 p: 55.. 3500.0 . $0000.90 000.. 55.5.. 0 . 50:51.05 00.10.. 5500.0 . 505200.505 03.50.. 300000.500 8.. 0.: .55... 3.00.0 .. .2023... 30. 2.5.. 0 0 505.501.: 0300.5 50508.0 . 0 0.00.000 “5555- 50.005 5.... 0: 53000.0 0 55 0.50-950 ”555- 55.055 5.3 05.5 5.0 05 5.0 05 0.0 55 5.05 .005 0.00 055 5.05 000 5.0 5 5.0 0 5.0 5 5.0 55 5.5 55 0.5 55 5.05 005 5.55 050 5.05 005 050505 55.05 0 0 0 0 0 0 0.50 n 0.50 5 5.55 0 0 0 0 0 0 0 5.05 5 5.05 5 0 0 0.50 n 0.50 n 0.50 0 205.0 .0 55.35 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.005 5 0.005 5 0.005 5 0 0 0 0 0 0 005005 .00 .0 50035 . . . . . . . . .n 055.2505: 5 05 5 0 0 0 0 0 05 n 5 00 0 5 00 0 0 0 0 0 0 0 5 05 5 5 05 5 0 0 5 05 5 5 05 5 5 5 5 5.00 205000» .0 555.05 0 o 0 . . 0 0 0 6.: a 0! GOU‘I! 0 0 0 0 0 0 0 an 5 0 05 5 5 0n 5 0 0 0 0 0 0 5 5 5 5 5 5 5 n5 5 5 05 m m 05 5 50000520 .5 5.05-c5 . . . . . . . . . . . . . .20 0.50 0 5 5 0 5 5 0 0 5 55 505 5 50 555 0 05 555 0 0 m 0 5 0 0 5 5 0 05 5 0 55 5 5 05 5 55 05 5 55 00 0 55 55 5.0052 .5 55555.05 0.0 5 5.0 0 0.5 55 5.05 00. 5.50 555 5.00 500 5.0 5 5.0 0 5.0 5 5.0 55 5.5 55 0.5 50 5.05 005 5.55 505 0.05 .50 205Huuunw .5 0 .02 o .02 o .02 0 .02 5 .02 o .02 0 .02 o .02 5 .02 0 .02 0 .02 o .02 0 .02 o .02 o .02 25:5 5.0.. 5.3 .5552 5.02 :3 .55... 002 :3 .555: 5.0: :3 55552 002 :3 .5520 055 052 052 050 050 055 055 052 050 055 055 055 055 052 055 05025050005 555-» .5055: 5.03m 532.5 30505:» .955. .56 53555500»:— 5.55.. 5.5-5.5.2503 56 one. no 5.35.52.50.63...— 05552.5. 65 .50.0 5 5 .555.5 u 5x 5 ~ .50.0 A 5 .550.0 a 5x 555: 055 .50.0 A 5 .050.0 . m5 "002 055 .50.0 v 5 .500.5 u m5 555500 55.505 005: 055 .50.0 v 5 .500.5 . “5 "55500 50.005 002 055 5 "205 055 .50.0 5 5 .005.0 . “x ”55500 55.055 205 055 .moncoa50¢ 95L 5:00550i< opu—¢:< 5.55 050 0.05 055 0.50 550 5.05 0005 0.00 055 5.05 000 5.05 005 5.55 050 5.55 505 0.55 0005 5.00 005 0.00 550 550505 55.05 5.55 0 5.55 0 5.50 5 5.50 5 5.50 5 5.55 0 5.50 5 5.55 5 0.50 5 5.55 0 5.55 0 5.55 0 00550 .0 55.05 0.005 5 0.005 5 0.005 5 0 0 0 0 0 0 0 0 0 0 0 0 0.005 5 0.005 5 0.005 5 005005 .00 .5 50055 05cann5= 0.05 5 5.55 5 5.55 5 0.05 5 5.00 0 5.00 0 5.05 5 5.05 5 5.55 5 5.50 5 5.50 5 5.00 0 5.00 5550050 .0 555.05 .0< 5.00 0 5.00 0 5.50 5 0.55 5 0.55 5 5.05 5 5.05 5 5.05 5 5.05 5 5.50 0 5.50 0 5.50 0 0005502 50035.50 .5 5055.05 0.05 55 0.55 50 0.50 00 5.55 505 5.50 555 0.05 555 5.55 55 5.55 05 5.55 55 5.50 505 5.05 005 5.50 555 0V5muwu .5 55555.05 0 I I O 0 I I I O sflau=IU 5 55 505 5 05 000 5 50 555 5 05 050 5 50 555 5 05 500 5 55 005 5 55 555 5 55 050 5.05 500 0.50 550 0.00 555 505502 .5 5 .02 5 .02 5 .02 5 .02 5 .02 5 .02 5 .02 5 .02 5 .02 5 .02 5 .02 5 .02 5: 2552 002 205 2052 002 2:5 5552 002 505 2552 002 205 05:25mwwu05 0.55 0.55 055 0.55 0.55 0.55 05.5 0.55 0.55 0.55 0.55 0.5.5 hOuumncoz FOHV5Q nHOUMIUp—OZ nhOunflh. .mauaum 5ouu5ncozuuouu5n can 5ounuu=oz-50000b Uh: can ucoucoanoz «0 can: mo semuuanlooau.av~ amauh 66 percent for Blacks (difference 1.5 percent) while for the medium PTC concentration these values are 67.8 percent and 74.7 percent for a dif- ference of 6.9 percent. A 7.4 percent taster difference (74.9 percent versus 82.3 percent) for these racial groups is observed at the high PTC concentration. These differential frequencies are significant fbr both the medium and high concentrations of PTC (p < 0.05). Table 14b also shows that when comparing bitter-nonbitter PTC responses of Whites and Blacks, no significant differences are observed (p > 0.05) for any of the PTC concentrations. Taste Perceptions and Smoking:5tatus of Respondents Data comparing the taste perceptions of control solutions fOr smokers and nonsmokers are compiled in Table 153. The nonsmokers who constituted the majority of subjects surveyed (82.1 percent) were found to misclassify the sour and sweet controls with slightly greater fre- quency while smokers tended to misperceive the tasteless, salty and bitter controls more often. That these small differences were not sig- nificant can be seen from results presented in Table 15b in which error categories are compared. The overall predilections fbr "correct" or "incorrect" classification of controls are similar in both smokers and nonsmokers with no significant error associations observed for these groups as evidenced by the Cramer's V and Lambda values. A similar lack of significant difference of taste perceptions of Antidesma between smokers and nonsmokers can be noted from data tabulated in Table 16a. Differences between these groups ranged from 0.1 to 1.0 percent for Antidesma I and O to 3.5 percent for Antidesma II. Such differences when analyzed by statistical tests show no 67 v.— ON o.wm m~v~ 0.9 mm v.nm new— m.~ on m.hm Nov— a._ nw ~.ma -v~ m.h— th _.~o —w- unuOP .om__.=u 5._ ON m.n¢ oo__ «.0 on o.nm co- c.~ mN 0.5m Nms. e._ h_ a.mm no__ ~.a~ m_~ o._m moa whosoemeoz Aam~acs o o o.oc~ mmN v.“ o. o.~m mmN _.n a m.om omN o.n o. _.oa cam n.o_ Ne h.nu o_~ muoxoem a .oz » .oz a .oz ' .oz * .oz » .oz « .oz a .oz a .oz v .oz woof—Our: MUOHMOU HOOP—Cup; uUthOU yup—Moor; uuohhou uUOtOU—um HUOHBOU uUflthUF—H HOP—MOD pouucou uooxm ~ouucou nouuum ~ouucou xu_am _ouucou awe—cumah mouucoo usom .myouucou mo coflunouuoa we: ucovconuou mo nauaum ucwxoew mo confiuansou--.um~ o—pah 68 Table 15b.--Comparison of Smoking Status of Respondent and Misclassifi- cation of Controls. Count Row % Smoking Status Row Column % Total Total % Smoker Nonsmoker 192 900 17.6 82.4 1092 0 74.4 76.3 75.9 13.4 62.6 54 214 Errors 1 20.1 79.9 268 20.9 18.1 18.6 3.8 14.9 12 66 15.4 84.6 78 .i 2 4.7 5 6 5.4 0.8 4.6 Column 258 1180 1438 Total 17.9 82.1 100.0 Cramer's V = 0.03040 Lambda (Asymmetric) 0 with Error dependent 0 with Smoking Status dependent O Lambda (Symmetric) (59 o . Auuuuoesxmv «vases acovcomoc _~ «Emoun:< sag: o I .ucovcunuv nauuum weaselm gum: c I aomuuoslxm m.uoeauu o . Ausuaoaaxm. «cases ueovcoaov u osmokuc< and: o n .acoecomoo nauaum ucwxoem no“) o a nuwuuoeaxm n.uoauuo “a namoeflu=< uncouom cu ~.c m.o o.” o.c o.~ ~.c m.n o._ c v.o noucouocuua o.o a eo.o a «.mN woe n.__ no. v.m~ mom m.cm Nah a.mv coo o.on n~m o o ¢.o - _aaop aoo-.:v h.o o o.o a m.n~ onn a.__ an_ a.v~ «oN a.om _oo m.ov oem a.mn NNv o o m.c a “yoga-ucoz Aam~.=. o o o o m.m~ ea ~.- mN m.u~ _n m.om _n_ a.mv ___ m.om mm c o ~.~ n «Logo-m a .oz » .02 a .oz a a .02 , .oz a .82 a .oz a .82 o .oz _~ u< _ u< __ e< _ e< __ u< _ v< __ 1< u u< __ u< . e< Auunm nouuwa Hausa uaom unoaounnh .uanova»:< we acumumouuom sud: ucovcommoz mo unuaum acuxolm mo :ouuuamlou-c.ao~ o—ach 70 significant associations between Antidesma perceptions and smoking status. Analogously, when these perceptions are compared with respect to the bitter-nonbitter classification as in Table 16b, smoker-nonsmoker differ- ences were also insignificant (p > 0.05). In Table 173, comparisons of smoking status and perceptions of PTC are reported. Consistent differences between these groups were observed fbr the majority of taste responses. As shown, for all concen- trations of PTC, smokers were more likely to find these solutions taste- less and less likely to perceive these as bitter or sour than were the nonsmokers. For sweet and salty perceptions (smallest categories) no trends could be discerned. Despite the apparent unifbrmity of taste perceptual differences for each of the five perceptions reported, no significant overall associations of any PTC concentration with smoking status could be confirmed. As reported above, smokers more often judged each PTC solution as tasteless when compared to nonsmokers. When the absolute frequencies of tasters and nontasters are compared as in Table 17b, greater differ- ences were found between smokers and nonsmokers fer the low concentra- tion of PTC. This difference was significant at the 5 percent level. For other PTC concentrations, no significant differences were observed between these two groups. When PTC bitter and nonbitter responses by smoking status are compared (Table 17b), as expected from previously discussed results, the frequencies of nonsmoker bitter responders were greater at each concentration than those of smokers. None of the dif- ferences however, were significant for any of the PTC solutions tested. 71 Table l6b.--Comparison of Smoking Status of Respondent and Antidesma Bitter-Nonbitter Status. Antidesma I Antidesma 11 Smoking Bitter Nonbitter Bitter Nonbitter Status No. % No. % No. % No. % Smoker 29 11.2 229 88.8 76 29.5 182 70.5 (n=258) Nonsmoker 139 11.8 1041 88.2 329 27.9 851 72.1 (n=1180) Total 168 11.7 1270 88.3 405 28.2 1033 71.8 2 Xi = 0,061, p > 0.05 x1 = 0.26, p > 0.05 Taste Perceptions and Time of Last Food Eaten (Elapsed Time) By Respondents In an effort to discover if relationships exist between taste perceptions and the time of last fbod eaten by subjects prior to taste sampling, respondents were grouped into seven elapsed time categories: (1) 0.1-0.9 hours, (2) 1.0-1.9 hours, (3) 2.0-2.9 hours, (4) 3.0-3.9 hours, (5) 4.0-4.9 hours, (6) 5.0-9.9 hours and (7) 10.0 hours or more. These analyses included 1,425 subjects since thirteen individuals failed to record the time of the last food eaten. When elapsed time is com- pared with respect to perceptions of controls (Table 18a), frequencies of misperceptions of these solutions are comparable throughout each elapsed time category. When the misclassification error rates were analyzed by these elapsed time groupings no significant associations were uncovered (see Table 18b). 72 .o - Retain. .3..— .o . 3.213 .33.. 5.31.2.8 3. at f... a .2923... Logo-m 5... a . 3.2812. .3]: £522.... :3 u: 5... a $51.82. 8.8m 5... o . 3.3815: .38.. .o a stain. Iva-1. .237.er 5a.: EL 5:) e .ucovcoaov .3:an so.) a I canals. .33 "mmo'oé I > 3.3.50 u:\-I «NJ: 8.3.. P: gunned o > ulihu ”:3 no.0: v! P: "2'86 0 > «lot-bu ":qu :6: 33 U: «.0 2 «6 2 .6 2 72. .02 0.3 So .6... con 86 _ n... m To a; Z a.» an -.n an ~.v~ .3 n.—n on. ~.on no... 233. .6 e. .6 a .6 2 n .2 on. n .3 o2. 0.3 no no... _ n .o v on... _ ed 3. ad 0v .4 o. «.3 o2 s6» 3n gin a: 31328.. o a v6 — .6 ~ 0.00 2.. «.3 3— 0.3 an. o o v... _ .6 s6 2 5 ~ 5 n n n v.3 3 73 o. 0.: a: ~3on o .8. o .8. o .3. o .8. 0 .oz 0 .3. a .02 a o a 62 0 oz 0 oz 0 .9: o .0: o .9. i... 3. .3 .3: 3. .3 a... 2.. .3 5:: .5. .3 a... a... :3 a..." :28» 0.: PE PE PE HE E EL PE uh. E 8... E... PE 8... Pr. 3.2;, 53:- 32“ Son :23...— .PE «0 £253.35 ‘2: 2.01.3.3: uo nae-um 2.1..» no gingiin— 03-h 73 mc.c A a .ma_.o . mu "A~\u- m~._au sum: up; mo.o A a .mnn.e . ”x "fi_\u- no.o¢u no: up; no.9 A a .cam.~ . Mu "n_\uu ~n.c~u so; up; mo.o A a .oa.o . M x “_xa- m~.~av sum: up; mo.° A a .o..~ . “x un_\u- no.ovv no: we; me.o v a .om.» . _x A-\ua _n.o~ 38; pen a a.m~ on. o.on ._m a.m. ~no _.on noo_ o.vo c~o _.om can ~..~ men n._n one ~.an new a.ma coo" h.ao «as a.oo man _auoh Acc._.:. s.o~ can a.mn "a. a.m. hem n.8h can n.qo awn c.5m who a.n~ cm~ s.oa «on c.nn we. ”.05 com n.ao «_o c.~o Nab h3.8302 Acm~ucv o._n on c.5n no a.m. m~_ o.ao us~ ..~o _o. o._m nn_ ..o~ no ~.vn co o.ce m- o.nh oa~ a.mo ass v.mm a." noxoam a .02 9 .oz a .oz 9 .oz 9 .oz a .oz 9 .02 « .oz 9 .oz . .02 a .oz a .oz .3: as. :3 :3: no: :3 :3: co: :3 :3: to: :3 .33» p: p: p: or. p: at PE or. p: p: p: or. «5me .3: 3.32 .3qu 333 «:02 33nd... .333 “033.5233qu EB uoungucozéounc... PE .33 ace—3.0.33. uo nae-um med—elm mo cent-9802...:— 03:. .mwe—u: .coumo voOu um- mo 05“» covuouou on: maoomasm mo nanxgac m.uoemao o.oo~ a.mH «.0 ~.v o.o~ e.wfl N.8N m.mH s swoop mNQH BNN mm so Hms Now «an meN season w.o N.o N.o o.o M.H N.H m.o m.m e.m o.m o.o m.o m.¢ 5.4 mm. mm o.e~ o.v o.e a.mfi o.e~ N.NN m.n~ NH m m m as as me m.m H.H w.o ~.~ m.~ m.e m.~ a.mfl a.mm «.mH o.o~ ~.H~ a.mH «.wfl a.mH H moa e.o~ o.o m.e H.~H m.m~ o.e~ m.mH em as NH mm He me He m.- w.v ~.m a.“ ~.v~ ~.o~ 5.4H H.85 m.os a.mh a.ma a.mh m.h~ c.5n m.mh o mmofi w.e~ v.0 H.v H.o_ n.m~ m.o~ m.m~ so“ me me oHH mom mww mom «HOHHm U d O +o.o~ m.m-o.m m.e-o.e m.n-c.m m.~-o.~ a.m-o.~ m.o-H.o “waaehsmow sauce 30 mmnno :HV oEH mam pom 30: m = . .e um Hm assoc .mHOHucou mo amnouumv :oflpmowmfimmufiomwz van flucovconmom An coumm boom umma mo oEwHV oswh vommmfim mo comflnmmEOUI-.nw~ manah 76 Specific taste perceptions of Antidesma solutions with respect to elapsed time are presented in Table 19a. As with perceptions of con- trols, no apparent elapsed time trends are presented for any of these perceptions fbr Antidesma 1 or Antidesma 11. Thus, time of last food eaten is not significantly associated with Antidesma perceptions (see Cramer's V and Lambda values). When bitter-nonbitter perceptions of Antidesma are compared by elapsed time a similar lack of perception-elapsed time trend is observed (Table 19b). Although the highest frequencies of the bitter response was seen at 4.0-4.9 hours after last fOOd eaten for both Antidesma solutions, this difference as well as overall bitter—nonbitter differences by elapsed time were not significant. Similar tabulations of effects of elapsed time on perceptions of PTC are reported in Table 20a. For each PTC concentration, the pro- portions of subjects who judged these solutions as tasteless appears to decline up to three hours while the proportion reporting bitter responses increase up to this same time. In subsequent elapsed time categories, frequencies of subjects reporting these perceptions appear to fluctuate randomly as do those of subjects who recorded other per- ceptions (sour, sweet and salty) across all elapsed time categories. Analyses of these overall responses reveal no significant differences with respect to elapsed time. PTC perceptions by elapsed time were also analyzed in terms of the taster-nontaster and bitter-nonbitter classification. These data are reported in Table 20b and show the same trends for the nontasters and bitter responders as previously mentioned. When these responses are analyzed by overall elapsed time groupings, differences between 77 o . NuNeuosaxmv «asses acovconou __ mamovmuc< saw: o u acoucoqov osmh vommu_m so“: a n Aowuuoesxn n.uosauu n- memovquc< o u Aowuuossxmv apnea; acovcoaov u «Emovmu:< sum: o - acovconoc asap venue—m sud: o u Aofluuossxm m.uosuuu "u numovuuc< 0.: a o.o a _.NN _ce a.__ ao_ N.mN awn g.om vNN _.o¢ Nmo o.on n_m c o m.o NN “_Auop NNNNucv o c o c N.NN co m.N_ oN o.nn mN m.mm oN~ a.mn mm n.~n ~N o o a.N 4 .ng: .o.o. RNU Num.:u N._ . _.. . v.oN mN m.N~ ~_ _.oN NN N.Nv N. n.vv an o.on «N c o c a .ma; a.m-o.m Nov Noo.=v N._ _ o o N.~n m_ N.o. o. N.oN oN N.No Nn o.ov QN N.NN n_ a a o c .Aag a.v-o.v Ame ANA—.eu c o a c N.NN _v 8.N n_ N.mN an o.Nc _N N.Nv NN ¢.vc No a o a c .Aa: a.»-o.n Ave NNoNueu e.o _ o.c N m.NN NN v.c NN N.NN No a.mm Nm_ m.oe NN~ ..Nn mm c o v.° a .nu: o.N-o.N Nag N.NN-cg m.c N n.~ m N.oN co. _.n_ a. o.n~ on N.Nv Na_ N.ov o¢_ a.mn va~ o o _.s c .uug a.N-o._ NNV anoNucc N._ o o o v.on an _.v. Nn ..NN an n.nv '__ °.ov N" c... me. o o _.N n .Au: o.c-_.o NNV N .oz A .oz A .oz 9 .oz A .oz a .o: a .oz a .02 a .02 p .oz - u< _ u< _. e< N e< a. v< _ e< __ u< _ u< NN e< N u< and? venue—u au—um nouuua «003w know mac—cunah .almovauc< mo meowuaoouom so“: Aucoveommoz an enema noon and; no onwhv olwh conneum no comwuanlou.-.om~ oNnuh 78 mo.o A a .vm.H u wx ANN mammeNuc< mo.o A a .mmN.N u ox AH mammefiue< N m.NN «Neg N.wN Hoe N.NN NmNs N.NN moN mNaooe nNNNueu N.NN noN N.NN we m.wm HON m.HH 0N .mp: +o.oN .N Amwuev o.NN no v.wN mN m.Nm NN m.NN NH .mn: m.m-o.m .8 Agency n.wo He N.NN as a.mm om N.oN oN .mng m.e-o.e .m NHmNuev N.NN oNN N.NN “e v.Hm mm“ o.m mN .mu: m.m-o.m .e NNoNucv m.NN omN m.NN NN 8.Nm OVN a.” NN .mu; 8.N-o.N .m fivaucv N.NN «NN N.oN ooN m.ow mNm N.NN me .mp; m.N-o.N .N AneNucv a.mo mwN ¢.om om a.mw eNN H.4N Nm .mug m.o-N.o .N A .oz N .oz N .02 N .02 Houuwncoz nouuwm nouuwncoz nouuwm oefih vommmfim NH «amoeNu:< N mammeNue< .mSHmpm Houaflncoz -uouuwm mamovfiuc< nu“: aucovcommom xn :oumm boom ummg mo oefihv QEMH vomamfim mo :omfinmmEOUuu.an ONAMH 79 .Ncoacoaos NN_= use g... o . Nu...u-NNN .ua-IN “Ncoacoauq oa.N N.NN... NNN: Noooc.o . NuNNN.-s. a.m.-1L9 N:\ul no.0: 3 2L .Ncougcete an. up; NNN: NNNNe.e . NuNLNo-INNN can-IN Nucuucoaos «INN no.5..u g... N.Nao.o . NuNNNo-usn .50: 25 no...» «mo.— "oa: woman; :3: «33m 333—52383; can nounqucozfiouma... PE .3 :omNNaN—Iou--.oo~ 03a... mmnoo.c . NuNNNoeame «unsuN NeueeoNoe NNN UNN NNN: a . NcovcoNou NN NamouNNN< NNN: n.0eo.o . NoNNNoa-Nn N.NonNNu "NNN u N.NoaaNu NNN u N.Noaato "NN .amouNN=< .c - NoNNNQIINmN avaauN NueoveoNoo nsto: UNN NNN: o . NcoucuNou N u< NNN: o . NuNNNoa-NN N.Noaauu "N NaNouNNc< o.o a 0.9 m N.NN mo. N.NN ooN v.m~ men N.NN NnN a.mv coo o.cn NNm o a.c NN NNNNOH NNN.NN c o o o o.om m o.cN N o.oN N o.om m o.o. v o.oe v o o o NNNaw Aommucu v.c v o.o N n.m~ nn~ N.NN NcN N.o~ ov~ N.om Nov N.Nv nvv n.Nn new 0 o.o N noNNNn Nana. o c o o o.ov N o.c~ N o c o.o. m o.oo n o.ov N o o o Nooxm Nnm.=N m.— H o o N.Nv m~ o.NN m o.NN a N.Nv oN o..n oN N.Nn NN c m.N N Naom HomvueN 5.: n v.o ~ N.Nn ovN N.NN mm o.m~ mNN N.Nm mn~ N.~v NNN o.nn NmH o m.o v unoNounuh N .02 N .02 N .oz N .02 N .02 N. .02 N .02 a .02 .02 N .02 NN v< N v< NN v< H v< NN v< N v< NH v< N u< NN v< H u< Hquamo.ch Istu: uhN NNNom soNNNn Nooxm Naom nmoNonah .usmotNN:< No mcoNuNouuoN gum: ukN mo :oNNNNNcoueoo astox mo mum:0Nmo¢ oumuh mo :OnNuuNlou--.aN~ oanh 89 mo.o A N .NNN.N u wx mo.o A N ..NN.N u wx N.NN NNN o..o. NNN N.NN om. N.NN NNN mNNNoN NNucN N.NN N N.NN N N.NN N N.NN N NNNmm NNoNuNN N.NN No N.NN NNN N.NN mm N.NN NNN NNNNNN NNmNueN N.NN NNN N.NN No. N.NN NNN N.NN .N. Noozm NNNmucN N.NN .NN N.NN NNN N.NN NNN N.NN com Naom NNNuNN N.N. m N.NN N N.NN . N.NN N mmoNonNN N NamewNNN< N .oz N .oz N .oz N .oz Nouuwncoz DEN NouNHm DEN muoummucoz DEN mpoummE DEN .mzumum Houancoz uuouufim van Noumau:02unoummE estoz DEN ANN: mow=0NmoN HH vcm H «Emowwu:< mo comwumNEODnn.n- oHnmE mo.o v N .Nom.NN u Mx No.0 A N .Nmo.m u Mx N.NN NNN N.NN NNN N.NN om. N.NN NNN NNNNNN NNucN N.NN . N.NN . N.NN N N.NN m NNNNm NNNNuNN N.N. NNN N.NN NNN N.NN NNN N.NN NNN NoNNNN NmomucN N..m NNN N.NN o.N N.NN NNN N.NN NNN Noozm Noooucv N.NN NNN N.NN m.. N.NN NNN N.NN No. Nsom NN NENNNNNN< N .oz N .02 N .02 N .oz nouuwncoz DEN Houuwm DEN mpoummucoz DEN mnoummE DEN .eosaNpcou--.NNN NNNNN 91 significant difference was fbund fbr overall Antidesma II perceptions with respect to the PTC bitter-nonbitter dichotomy. Similar results were obtained when these same PTC classes were compared to bitter- nonbitter groupings of Antidesma as in Table 22c. As shown, statis- tical significance was observed only when Antidesma II bitter-nonbitter perceptions are contrasted with bitter-nonbitter responses to this medium PTC concentration. For the high concentration of PTC, similar trends as reported for the previous two concentrations are evident, in that overall per- ceptions of the Antidesma solutions are not significantly associated with PTC responses (see Table 23a). Application of the PTC taster- nontaster and bitter-nonbitter classifications with individual percep- tions to Antidesma I and II (Table 23b) results in significance only fOr the PTC bitter-nonbitter category in Antidesma II. Furthermore, as with the previous PTC concentrations, comparisons of bitter—nonbitter per- ceptions of Antidesma solutions with the high PTC concentration (Table 23c), one finds significant differences only with respect to PTC bitter-nonbitter responses for Antidesma II. It should be noted that in all of the above instances where significant differences were found, such differences were primarily due to the less than expected frequencies of subjects who responded bitter to Antidesma as well as bitter to PTC. Comparisons of Taste Perceptions of Antidesma I and Antidesma II Perceptions of the two Antidesma solutions are compared in Table 24a. As is apparent, the major combinations of responses were sour responses to both which were recorded by 324 subjects (22.5 92 mo.o v N .NNN.NN u Nx mo.o A N .NNN.N u Nx N N N.NN NNN o..o NNN N.NN om. N.NN NNN NNNNNN NmmoNuNN N.NN N.N N.NN NNN N.NN NNN N.NN NNN NNNNNNNoz Nmo.u:D m.N. NNN N.NN NNN N.NN N.N ..me NNN NNNNNN HH «Emouwu:< mo.o A N .Nco.o u Mx mo.o A N .NNN.o u wx N.NN NNN N.NN NNN N.NN om. N.NN NNN mNmNoN NNNNNueN N.NN Nm. ...N NNN N.NN NNN N.NN NNN NoNNNNNoz NNNNnNN N.NN No N.NN NNN N.NN mm N.NN NNN NNNNNN N mamocNNN< * .oz * .02 w .02 w .02 NNNNNNNoz oNN NNNNNN UNN NuNmmucoz NNN NNNNNN oNN .mauwum Houufincozuuouuwm mam Hoummu:02unoummE suave: DEN new: momcommom nouufincoz-nouuNm mamokuc< mo :omHHmNEODuu.omm oHnmE 93 a eNco.o n NNNNNNNNN NNN; NNN NNN: o . .NceuceNuc NN cN NNN: NNNNN.N a NNNNNoast N.NoaaNu "NNN N N.NoeNNu NNN NNN N NNNNNNNc< Homuuoaaxwu apnea; NuNNNosaNNN NNNNNN N.N N o.o N N.NN Ne. N.NN NNN ..NN Non N.NN NNN N.N. coo N.NN NNN N o N.N NN NNNNoN NNN-ND N N N N N.NN N N.NN N N.NN N N.NN N N.NN N N.NN N N N N o NNNNN NNNNNucN N.N N N.N N N.NN NNN N.NN NNN N.NN NNN N.N. NNN N.N. NN. N.NN NNN N o N.o N NoNNNN NN.=N o N o N N N N o N.NNN N N o N N N.NNN N N N N N Noozm NNN-ND N N N.N N N.N. on N..N .N N..N NN N.N. NN N.N. NN ..NN NN N N N.N N Naow NN.N.=N N.o N N.N N N.NN NNN N.NN o. N.NN NN N.NN NNN N.o. N.N N.NN NNN N N N.N N ANoNoNNaN N .02 N .oz N .oz N .oz N .oz N .oz N .oz N .oz N .02 N .oz N a. N NN NN e. a. NN v. N NN NN a. N a. NN u< N N< NNNN- NN.NNN NNN: NNN NNNam NouuNn Nooxm Naom mmoHounaE .asmoeNN:< mo mcoNuNouNoN :NN: DEN we :oNuuNucoucoo :uNz mo momconoN oumaE mo :oNHNaNIODuu.an~ oNnuE 94 NN.N A N .NN.N u mx NN.N A N .NNN.N u Mx N.NN NNN N.NN NNNN N.NN NNN N.NN NNNN NNNNNN NNuNN N.NN N N.NN N N.NN N N.NN N NNNNN fiNNNncN N.NN NN N.NN NNN N.NN NN N.NN NNN NNNNNN NNNNuNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN Noozm nNNNuNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN Naom NNNuNN N.NN N N.NN N N.N N NN.NN. NN NNNNNNNNN H asmokun< N .oz N .oz .oz N .oz uouuflncoz chm hmuuwm up; muoummucoz up; muoumme up; .mzumum Noupwncoz -uouuwm van nopmmu:02uuoummh saw: Ohm spa: mmmcommom NH vcm H «Emovfiu:< mo :omwnmmEOUuu.nmm munah 95 NN.N v N .NNN.NN n Mx NN.N A N .NNN.N u Nx N N.NN NNN N.NN NNNN N.NN NNN N.NN NNNN NNNNNN NNuNN N.NN N N.NN N N.NN N N.NN N NNNNN NmNNuNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN NNNNNN NmNNuNN N.NN NNN N.NN NNN N.NN NN N.NN NNN Noozm NNNNuNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN Naom HH mammgug N .02 N .02 N .02 N .oz NNNNNpcoz NNN NNNNNN NNN NNNNNNNNNZ NNN NNNNNNN NNN .NNNNNNNoN--.NNN NNNNN 96 NN.N v N .NNN.NN u Mx NN.N A N .NNN.N u Mx N.NN NNN N.NN NNNN N.NN NNN N.NN NNNN NNNNNN NNNNNuNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN NNNNNNcoz NmNNuNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN NNNNNN NN NNNNNNNN< llllllllllllllllll AT.n......|.|.|.|.:.|.|.|.|.|.n.|.|.|.|.|.|.|.|.|.|.|.|.|.u.|. NN.N A N .NNN.N u Wx NN.N A N .NNN.N u mx N.NN NNN N.NN NNNN N.NN NNN N.NN NNNN NNNNNN NNNNNnNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN NNNNNNNoz NNNNuNN N.NN NN N.NN NNN N.NN NN N.NN NNN NNNNNN H «Emouwu:< N .oz N .02 N .oz N .oz nouuwncoz DEN pouufim DEN noummucoz DEN noummE DEN .mnumum nouan:OZuuouuHm van umummu:02uuoummE swam DEN zqu momcommwm Houufin:02upouuHm «EmouHu:< mo :OmHnmNsounn.unN mHnmE 97 N.NN N. N.N N.N N.N m.mH o.mm m.m w.w no N N. N.NN N.NN N.NN NN.N NNN N NN NN NN N.NN N. N.NN N.NN N.NN o.m~ ~.mv a.mN m.ov pmmzm N. N.NN N.NN N.NN NNN N NNN NNN NNN N.NN N. N.N N.N N.NN o.om H.Hm m.NH N.Nv Ndom N. N.NN N.NN N.NN NNN N NNN NN NNN w. o m. #0 v. N N.N N. N. mmO 0 mm N N.NN N.NN N.NN N N N NN N N N N mcoflumoonN N NsNoNNNN< NNN NNN % .m HO H 003 H30 NNNoN NN N NN.N N N N NNN Nou 3oz NNN 3oz mcoNuNmuumN HH mamovHuc< ucaou .HH mENovac< wad H mamovHuc< mo NaoNuNNUNmN ouNNE mo :omHuszounu.mv~ oHnmE 98 acovcmNov HH «EmmuNu:< :NNz mHomo.o acmwcmNov H mENocNuc< nufiz mmmno.o NNNNN.N u NNNNNNNNNNN NNNNNN HUNNNNEENNNN NNNNNN mowH~.o u > mNHoEMHD o.ooH c. ~.wm «.mm a.mv N HNHOE wmvH m mow mmm coo cesHOD o. H. N. N. H. m.~H N. w. m. N m H.HH m.mm m.mm N.NN NH m m H m m N mcoNuNooNoN H mammkuq< HmuoE NuHmm umuuwm poozm Hsom NW“ WWW u so 30m mcoHuNmopoN HH «Emovfiu:< a ”snow .NNNNNNNoN--.NNN NNNNN 99 percent of total sample). The second most frequent response combination was sweet perceptions to both (19.3 percent) followed by sweet Ad I-- sour Ad II (18.8 percent), sweet Ad I-bitter Ad II (12.7 percent), sour Ad I-bitter Ad II (8.8 percent) and bitter fer both solutions (6.2 per- cent). Other combinations were reported with frequencies of less than 4.5 percent. While these responses to Antidesma I and Antidesma II are not perfectly correlated, Cramer's V and Lambda values are much larger than those previously encountered in the PTC comparisons thus suggesting a stronger association between the overall perceptions of these two solutions than for PTC. This association is further substanti- ated by data presented in Table 24b. Comparisons of the major percep- tions of Antidesma I (sour, sweet and bitter) with the dichotomous fre- quencies of Antidesma II taste perceptions shows that in each case, these values are much greater than other frequencies obtained in that specific column. For example, the frequency of bitter perceptions fer both Antidesma solutions is 53.0 percent as compared to frequencies of 25-33.3 percent for other Antidesma I-Antidesma II bitter combinations. As verified by Chi-square values, each of the differences for comparisons made in Table 24b are highly significant. Although not presented but as expected from these data, similar significant differences are obtained when each Antidesma II perception is compared with dichotomized Antidesma I perceptions (bitter-nonbitter, etc.). Furthermore, as seen in Table 24c, when dichotomous categories of the major perceptions for both solutions are compared, differences are again highly significant. 100 mo.o v N .ve.v~H u vx choNuNooNoN poozmcoz-uoozw HH Namoowuq< Sufi: H «EmooNu:< N mc.o v N .Nm.mm n “X choHuNoonoN Naom:02unsom HH mEmooHpc< nuflz H «EmooHu:< NN.N v N .NN.NN u mx NNNNNNNNNNNN NNNNNNNoz-NNNNNN NN NaNNNNNN< NNN: N NENNNNNN< N.NN NNNN N.NN NNN N.NN NNN N.NN NNN N.NN NNNN N.NN NNN NNNN NNNNoN N.NN N N.NN N N.NN N N.NN N N.NN N N.NN N N NNNNN N.NN NNN N.NN NN N.NN NNN N.NN NN N.NN NN N.NN NN NNN NNNNNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN N.NN NNN NNN NNNNN N.NN NNN N.NN NN N.NN NNN N.NN NNN N.NN NNN N.NN NNN NNN NaoN N.NN NN N.NN N N.NN N N.NN N N.NN N N.NN N NN NNNNNNNNN N .02 N .02 N .02 N .02 N .02 N .02 Noozmcoz Noozm asomcoz Naom popuNncoz noupam .oz mcowumoouoN . . H Namoowuc< NaoNuNooNoN HH msNooNuc< .NcoNuNounoN Huoo3m:02uuoozm .Nzom:02nusom .Nouuwncozsnouuflmv HH Namoowuc< saw: macauNooNoN Namoowuc< HHmno>o chowusHom «Emoowu:< mo NaoNuNoonoN oummE mo :omwnmNEoD-a.pv~ oHoNE 101 NN.N v N .NN.HN u Wx N.NN NNN N.NN NNN NNN Naomcoz N.NN NNN N.NN NNN NNN usom N .Nz N .02 haemcoz uaom N.NN NNNH N.NN NNN NNN“ Hmaop No.0 v a .N.Nm u Mx N.NN NNN N.NN NNN NNNH NNNNNNNNz N.NN NN N.NN mm NNN NNNNNN mm .02 0A. .02 .oz nouancoz nouuHm HNpOH H mammkuq< NH NamNNNNc< .m:OHHHQOHom Huomzm:02numozm .u30mcoz-p=om .umuancoZIHouuHmv HH mammuHuc< zqu mcoHumoouon nuomzm:02uuoozm .H30mcoz-hsom .umuancoz-NouuHmv H «EmovHu:< mo :ONHNNHEOUuu.ov~ memh 102 N.NN NNNN N.NN NNN NNNN NNNNN NN.N v a .NH.NNH u Mx N.NN NNN N.NN NN NNN Noozmcoz N.NN NNN N.NN NNN NNN poozm N .Nz N .02 poozmco umoz .02 H «Emu an: 2 m HNNNN N. < HH mammcHu=< .NNNNNNNNN--.NNN NNNNN 103 Taste Perception Family Studies To determine if Antidesma taste perceptions are consistent with a simple dominant-recessive genetic hypothesis, data were obtained from 115 families. As reported previously, these family studies included taste perceptions as well as general demographic infbrmation for 112 two generation and 3 three generation families. In the analyses which fbllow, two generation families will be considered separately from those of three generation families. Two Generation Families--General Demographic Data The 112 two generation family data included information from 443 subjects: 224 parents and 219 children fbr an average sibship of 1.955 children per family. Within the offspring group there were 119 (54.3 percent) males and 100 (45.7 percent) females. With respect to race 88 (78.6 percent) families were White/Caucasian and 24 (21.4 percent) were nonwhite matings. In this latter category there were 21 Black/Afro American, one Chicano and two families in which one parent was white while the other parent was of Chicano or Spanish American heritage. Paternal ages ranged from 27 to 64 years (mean = 43.54) and maternal ages ranged from 28 to 62 years (mean = 41.30). The overall parental mean age was 42.43 years. Age ranges for male and female offspring were 7-23 years and 7-45 years respectively with average male and female progeny ages of 13.86 and 13.31 years (Note: As previously indicated, offspring younger than age seven were not included in the sampling). As reported fbr the total sample, frequency data on smoking status and elapsed time since last food eaten were also obtained fbr families. Within the parental group, 62 (27.7 percent) were smokers 104 and 162 (72.3 percent) were nonsmokers. Among the offspring, there were 11 (5.0 percent) smokers and 208 (95.0 percent) nonsmokers. The time of last fOOd eaten by parents ranged from 0.1 to 15.3 hours with an average elapsed time of 2.25 hours since last food ingested. For offspring, this elapsed time range was 0.1-10.8 hours with a mean of 1.82 hours. Taste Perceptions of Families Data tabulated in Table 25a compares taste responses of parents and offspring fbr the control solutions. As can be seen, these data show that similar to previously reported results fOr the overall sample, parents and offspring more often misperceived the sour and bitter controls. Furthermore, the misclassification (error) rates fOr offSpring were slightly greater than those of the parental group for each of the controls except salty. Offspring also recorded higher average intensities for the control solutions. From the comparison of misclassifications of these solutions by parents and offSpring as reported in Table 25b, it will be noted that perceptual errors were not significantly different for the two groups (p > 0.05). Overall taste perceptions of Antidesma are recorded for parents and offspring in Table 263. Parents most often perceived Antidesma I as sour (50.9 percent) or sweet (30.4 percent). Conversely offspring perception frequencies were greater fOr sweet (43.8 percent) followed by sour (37.0 percent). Bitter perceptions of this solution were similar for the two groups (17.4 percent and 17.8 percent) while only individ— uals in the parental group judged this solution as tasteless and only offSpring reported salty perceptions. These differences in perceptions 105 me.m v.H m o.wm NHN ovm.m N.N m w.na mHN AuHmm nmo.v N.v w v.mm mow “Hw.m H.m n m.om BHN HoupHm moo.m m.o H m.ma mHN wa.~ v.o H a.mm mNN «603m nmo.v o.HH em a.mw mmH omm.m m.w aH m.Hm mom 930m moo.o m.o m H.mm NHN ooo.o c o o.ooH vmm mmoHoummH AuHmcoch N .02 N .02 qumcoucH N .02 N .02 :mo: uooupoocH poounou :moz poouuoo: uoopno m:0Hu=Hom : .o : : = H: = U: HORNE—CU mcHHANHHO mucohmm .mHopuaou mo mcoHumoUHom mcHnmmmmo cam Hanconmm mo :omHHmmEoUuu.mm~ oHnmb 106 Table 25b.-—Comparison of Misclassification (Errors) of Controls for Parents and Offspring. Frequencies of Misclassification (Errors) of Controls (Tasteless, Salty, Bitter, Sweet and Sour) Errors Parents Offspring No. % No. % 0 198 88.4 186 84.9 1 21 9.4 28 12.8 :_2 S 2.2 5 2 3 Total 224 100.0 219 100.0 x3 = 1.318, p > 0.05 of Antidesma I by parents and offspring were significant at the 5 per- cent level. For Antidesma 11 parents and offspring most often per- ceived this solution as sour, fbllowed by bitter and sweet with two individuals in both groups recording salty perceptions. Differences between the groups were not significant (p > 0.05). Furthermore, when bitter versus nonbitter Antidesma responses of parents and offspring are compared (Table 26b), no significant differences between the two groups were found fOr these taste perceptions. Table 27a compares the taste responses of the three concentra- tions of PTC for parents and offspring. Overall taste perceptions fre- quencies for these groups were similar for each concentration with most subjects judging these solutions as bitter or tasteless. As shown, Chi- square analyses revealed the lack of significant differences between the two groups fOr all of the PTC concentrations. Similarly, no 107 NN.N A a .NNN.N u wx NN.N v a .NNN.NH u Nx N o.ooH mHN o.ooH NNN o.ooH mHm o.ooH NNN mHmuoh N.N N N.N N N.N N N N NNNNN H.on co n.m~ mm N.NH mm N.NH mm Hoauwm a.mH we m.vH Hm w.mv Na «.om we woozm m.mv moH ~.Ho mmH o.nm Hm m.om vHH know o o o o o o m.H m mmoHoummH N .oz N .oz N .oz N .02 mcwummmwo museum; mcHnamwmo mucohmm HH mamoufiuc< H mamowwuc< .wcfinmmmmo was mucoumd How mEmouHuq< mo mcoHumoouom mamas mo comwummsoonn.mcm oHnmh 108 NN.N A N .NNN.N u wx NN.N A N .NNN.N u Nx NN NENNNNN=< N o.ooH mHN o.ooH vmm o.ooH mHN c.ooH NNN NHNHOH a.mo mmH m.on HNH N.NN owH N.NN me Nouuwncoz H.om co p.mm mm N.NH mm N.NH mm Houuwm N .oz N .oz N .oz N .oz wcwpmmmmo muconmN chnmmmwo mucoumN N NENNNNNN< .mcfinmmmmo can muconmN Now mmmcommom NouuNncoz-NouuNm mamowwuc< mo cemwummaou-u.no~ oHnmh 109 NN.N A N .NNN.N u mx NN.N A N .NNN.N u Nx NN.N A N .NNN.N u Nx N N o.ooH mHN o.ooH NNN o.ooH mHN c.ooH NNN o.ooH mHN o.ooH NNN mHmuoH N.N N N.N N N.N N N N N.N N N.N N NNNNN a.mo mvH m.oN mmH N.Ho NmH o.NN omH m.mm HNH m.oo mmH Houufim o o o o m.o H o o o o o o woozm N.N NH o.v m m.m NH N.N w N.N mH N.N o Haom N.NN em N.NN mm H.om co m.mN co N.Nm on N.Nm Hm mmmHoummH N .oz N .oz N .oz N .oz N .oz N .oz mcwummmmo mucoumN wcflnnmmmo mucopmN wcwpmmmmo muconN NNN: NNN NNz NNN NNN NNN .chhmmwmo can muconN pow UPN mo mcowumoopoN oummh mo :ONNNNNEoUnu.mNN oHnmb 110 significant differences were seen when parents and progeny were com- pared with respect to PTC taster-nontaster and bitter-nonbitter responses (Table 27b). Genetic Analyses of Antidesma and PTC Taste Perceptions From the taste perceptions for Antidesma and PTC recorded by individuals in families, 115 pedigrees were constructed (Fig. 5). As indicated, the top half of the pedigree symbols shows the individual perception of Antidesma I while the bottom half shows the perception of Antidesma II. Presumptive PTC genotypes are recorded below each symbol based on the responses recorded fer the PTC concentration of 81.25 mg/l and the general assumption that nontasting represents homozygosity for the recessive allele while the ability to taste PTC is determined by the presence of the dominant allele. Genetic Analysis of Family PTC Data (Two Generation Families) A summary of family PTC perceptions for the 112 two generation families with respect to types of matings and the resultant offspring are recorded in Table 28. When the numbers of offspring from the various mating types were tested fer randomness (chance) by Chi-square, it was found that the differences between observed and expected frequencies were highly significant (X2 = 80.89, p < 0.05). The data were then analyzed to determine if they confbrmed to the well established hypoth- esis that PTC tasting is dominant and nontasting is recessive. Follow- ing the estimate of q2 and q based on the total frequency of nontasters in the population sampled, Snyder's ratios were applied to calculate the proportions of nontaster offspring expected from various mating 111 NN.N A N .NNN.N n Mx NN.N A N .NNN.N u wx NN.N A N .NNN.N u Mx o.ooH mHN o.ooH NNN o.ooH mHN o.ooH NNN o.ooH mHN o.ooH NNN mHmuoh N.NN mm m.mN No a.mn mm a.mm vs N.NN mm N.mm mm NouuNncoz m.mo va m.on mmH N.NN va a.mo omH m.mm HNH m.oN mmH Houuwm NN.N A N .NNN.N u Mx NN.N A N .NNN.N u mx NN.N A N .NNN.N u Mx o.ooH mNN o.ooN NNN o.ooH mHN o.ooN NNN o.ooH mHN o.ocH NNN NHNNOH N.NN em N.NN mm H.om NN m.mN No N.NN NN N.Nm Hm mnoummucoz v.mN moH v.mN moH a.mo mmH m.oN mmH m.mo mvH N.NN mvH mnoumah N .oz N .oz N .02 N .02 N .oz N .oz mcfiummmwo mucoNaN chnmmmmo muconN chummmmo muconN NNN: NNN NNz NNN NNN NNN .wcfinmmmwo can mucoumN you UHN mo mcoNugooNoN NmuuNncoz-NouuNm New NopmmucoZunopmmH mo :omNNmQEou-u.nNN oHpmh 223 86552 n. N 0.23 .23... n N. 33w u ”moafiocmc PE 2:55.595 n 26.3 223.. 3:3 u _. 2.82.2 $sz u H No 2333.3 23... ”Eozom 1 .mEmmENNi Sow H No mcozamocwn. 2mm... " no... €5me 3223... u 8238 2:5“. 3233085359.; new mEmoEE< No 96:303.. 23... .o 3:55 02280 --.N N: t. 118 . _.m .om .aw 120 121 fl. ® . 2 3 w m .. _ .. 2 ...—. u... _ : EN. .5 N ’w" .L @N 2 z z 123 124 125 .3... S N» _ § ‘ C u... N... V\\s .5... CJ' .2: .9... 115. 126 127 mcwesmmm. mHmm.o u N.N u NN NHN>NmmoooN NH wcwummucoz mmcfipmz noummucoz x Noummucoz scum wcfinmmmwo Noummucoz mo :ofiuuomohm wouoomxm NNN.N u N.N u NN NNN.N m "mmcwumz noumwucoz x noummh Eonm mcwuammmo nmummucoz mo coNpNomoum wouoomxm NNNN.NN NNN+NN nmoNumN N.Ncwxcm kn. ”mwcwumz Noummb x Noummh scum wcflummmmo Noummucoz mo :oNuNomoum Noncomxm 0 "I'll- 1N mmoH o ov~.o .IMMWI. m o I. 0 IV 0 ImVVI 6 Nov o a NNN ox I van NNN o u mmw.: N nchNNmmmo mHN + mucouwm NNNU mvv u ammo many :« oHQEmm Hmuou oocNm moH chummmmo noummpcoz em + muconN nmummucoz mm u mm + AoHvN ”muoummpcoz mo xoconcmum Hmuou chm: N: mumeflumm ob "mcoNumHsonu mo mmHmENxm NN.N vv N .NN.NN u Nx N ”voucmmpr mama pom mmocsoccmm mo pmob .o>Nmmmoom mN mcfiummpcoz DEN NmNmonuomxc anon on mowumn N.NQNNNw :o woman; mHN em moH NHH Hmuoh oooo.H oooo.H HN HN o oH Hoummucoz x Hoummucoz mHmm.o mHHm.o Ho mH NV mm Noummpcoz x Noummk mmoH.o NNoH.o an NH mNH No Noammh x Hmummh «:oNuNomoNN :oNunomouN noummucoz noummh Nmummucoz Hoummucoz HmNOH .oz momxp chpmz Noncomxm uo>uomno chnmmmmo .mcowumooNoN umummpcoz-poumme UHN "NNNNNNN NNNNNN--.NN NNNNN 128 To test the difference between Observed and Expected Proportions of nontaster offspring from various mating types using the Z transformation: For Taster x Taster Matings: 22 = 0.1022 - 0.1099 3 _ 3.33;; = _0.2973 0.1022(1-0.1022) ' 137 Since Z is negative number: a/2 = 0.38591 thus a = 0.79182 F(z) Confidence in rejecting hypothesis = 1 - a = 1 - 0.79182 = 0.20818 For Taster x Nontaster Matings: z = 0.3115 - 0.3315 = _ 0.02 = _003373 1 V/’0.3115(1-0.3115) 0'0593 61 F(z) = 0/2 = 0.36693 thus a = 0.73386 Confidence in rejecting hypothesis = 1 - a = 1 - 0.73386 = 0.26614 For Nontaster x Nontaster Matings: Zo=._1_.-_1_.=0 1(1-12 21 F(z) = 0/2 = 0.5000 0 = 1.0 Confidence in rejecting hypothesis = 1 - a = 1.0 - 1.0 = 0 Combined Evidence - Conversion to x2 to give probability of accepting hypothesis: 2 x = - 2[Z 10ge a] 2 x - -2 [loge 0.792 + loge 0.0734 + 10ge 1] = -2 {-0.23319 - 0.30923 - 0] x2 = 1.08488* 0 p > 0.95 *Two degrees of freedom per a. 129 types (see examples of calculations). These were compared to observed proportions of nontaster offspring obtained by use the Z-transfbrmation to producecx values and subsequent statements of confidence in reject- ing the hypothesis (for explanation of this procedure, see Appendix). Evidence from each of the mating types and their offspring were com- bined and converted to a Chi-square value. As can be seen, the data are consistent with the hypothesis proposed (p > 0.95). Analysis of Family Antidesma Perception Data (Two Generation Families) Results obtained from dichotomous classifications of offspring of various matings fer each of the major taste perceptions of Antidesma (bitter versus nonbitter, sweet versus nonsweet, etc.) were analyzed by the same procedures used in analysis of the PTC data. In each case, initial analysis was performed to decide if data obtained were consis- tent with a random hypothesis then subsequently analyzed to determine if a dominant-recessive hypothesis could account fer observed results. For purposes of testing this genetic hypothesis, in each case the assumption was made that the basic taste perceptions (bitter, sweet, sour) were recessive. This was done because in the majority of cases, inspection of family pedigrees suggested that this assumption was the most feasible. Family data obtained for bitter-nonbitter perceptions of Antidesma I are reported in Table 29. The test of randomness fer observed and expected frequencies of these perceptions for the offSpring shows that the differences were significant (p < 0.05). Test of the genetic hypothesis that bitter perceptions of Antidesma I are recessive 130 N.N u NN u®>fimmvuwh mN nouuNn co>Nm mmcwumz NmuuNm x Houuwm Eonm chhmmwmo :Nouuwm: mo coauuomopm Newcomxm 3.: H ommN.o u lml.u m mchum: NouuNm x Noupflncoz Eoum chummmmo :Nouuwm: mo cofiuuomonm wouoomxm NNN+NN N NNNN.N u Illmll u N N mwcwpwz Houumncoz x NopuNncoz Eopm chpmmmmo :Nouuam: mo :oNunomopm wouoomxm NNNN.N u N .NNNN.N u NN o>Nmmooom mN :oflumoouod youuflm NmNmosuomx: oNuocmu mo amok NN.N v N .NN.NN u mx chNummwmo mo mcoNpmooNoN houuNQNOZuNouuNm mo mmocsovcmm mo umoh mHN mm owH NHH kuoh coco.H mnnn.o m N N m Houufim x Houuwm omm~.c omvN.o mm «H M? mm Hmuuwm x Houpwpcoz enmo.o ohHH.o mmH wH mmH wn Hmuuflncoz x Houuwncoz :oNuNomouN :oNuNomoNN Nouuwm NouuNncoz NmuuNm pouuNm Hmuoe .oz mmmxh mcwuwz voaovmxm uo>uomno mcfinmmmmo .mcowummoumm HouuNQNOZuNouuNm H mamokuq< ”moNv3um NHNEwmnu.mN oHan 131 To test the difference between Observed and Expected Proportions of "Bitter" offspring from various mating types using the Z transformation: For Nonbitter x Nonbitter Matings Z = 1.1615, F(z) = 0.87698 2 Since Z is positive number: 1 - F(z) = 0/2 = 1.0 - 0.87698 = 0.123 therefore a = 0.246 Confidence in rejecting hypothesis: 1.0 - a = l - 0.246 = 0.754 For Nonbitter x Bitter Matings 21 = -0.8772. sinxe z is negative number: F(z) = a/Z = 0.18943 thus a = 0.37886 Confidence in rejecting hypothesis: 1 - a 1 - 0.37886 = 0.62114 For Bitter x Bitter Matings Z0 = -1.6032 F(z) = a/z = 0.05480 thus a . 0.1096 1 - 0.1096 = 0.8904 Confidence in rejecting hypothesis: 1 - a Combined Evidence - Conversion to X2 to give probability that data conforms to proposed hypothesis: X2 = -2[2 loge a] = -2[loge 0.246 + loge 0.379 + loge 0.11] = -2[-1.40242 - 0.97022 - 2.20727] 2 X6 - 9.1598 0.5 > p > 0.1 132 and nonbitter is dominant resulted in a probability of 0.1 to 0.5 thus this hypothesis cannot be refuted. Results obtained for sweet-nonsweet and sour-nonsour perceptions of Antidesma I fer the various mating combinations and offspring pro- duced are tabulated in Tables 30 and 31. Although initial Chi-square analysis suggested that these results were random, genetic analyses were still perfbrmed. As expected, when such analyses were carried out, the combined evidence strongly suggested that these perceptions of Antidesma I do not conform to the proposed genetic hypothesis. Family data fer bitter-nonbitter, sweet-nonsweet and sour- nonsour perceptions of Antidesma II are reported in Tables 32-34. It can be seen that in each case, the test of randomness by Chi-square suggests that these data are not random (p < 0.05). The combined evi- dence from subsequent genetic analysis of the various mating types for each of these perceptions of Antidesma II however, strongly suggests that it is unlikely that they conferm to the dominant-recessive hypoth- esis proposed (p < 0.05). Analysis of Taste Perceptions from Three-Generation Family Data From Figure 5, it is apparent that families numbered 59, 62 and 63 include three generation taste perception data. Because of the limited number of these types of families and the absence of inferma- tion for several first generation members, these data were not con- ducive to detailed analyses. With respect to perceptions of PTC as can be seen, there were no exceptions observed which were inconsistent with the previously accepted hypothesis of dominance fer PTC tasting and recessivity for 1253 Table 30.--Family Studies: Antidesma I SweetINonsweet Perceptions. Offspring Observed Expected Mating Types No. Total Sweet Sweet Nonsweet Sweet Proportion Preportion Nonsweet x Nonsweet SS 67 42 109 0.3853 0.1431 Nonsweet x Sweet 46 48 43 91 0.4725 0.3783 Sweet x Sweet 11 8 11 19 0.5789 1.0000 Total 112 123 96 219 Test of Randomness of Sweet-Nonsweet Perceptions of Offspring: x3 - 3.2049, p > 0.05 Test of Genetic Hypothesis: Sweet Perception is Recessive 2 q I 0.3702 q I 0.6084 52 I 0.1431 S1 I 0.3783 50 I 1.0000 For Nonsweet x Nonsweet Matings 22 I 5.197 F(Z) I 0.99999997133 a Z 0 Confidence in rejecting hypothesis 2 1.0 For Nonsweet x Sweet Matings 21 I 1.801 F(Z) I 0.96407 0 I 0.0718 Confidence in rejecting hypothesis I 0.9282 For Sweet x Sweet Matings Z I ~3.72 F(Z) I 0.00010 0 I 0.0002 0 Confidence in rejecting hypothesis I 0.9998 Combined Evidence - Conversion to x2 to give probability that Antidesma I Sweet perceptions conform to recessive hypothesis: x: - 33.2621, p << 0.005 Note: This Chi-square value and others like it is an approximation since natural logs for two a values could not be determined from published natural logarithm tables, loge 0 I I and loge 0.001 I -6.90776 Thus for values of a I 0 and o I 0.0002, natural log value used in Chi-square calculation was -7.0 for both of these a values. 1254 Table 31.--Family Studies: Antidesma I Sour-Nonsour Perceptions. Offspring Proportion Proportion Hating Types No. Total of Sour of Sour Nonsour Sour Observed Expected Nonsour x Nonsour 29 39 13 52 0.25 0.159 Nonsour x Sour 52 63 43 106 0.4057 0.399 Sour x Sour 31 36 25 61 0.4098 1.000 Total 112 138 81 '219 Test of Randomness of Sour-Nonsour Perceptions of Offspring: x; - 4.2031, p > 0.05 Test of Genetic Hypothesis: Sour Perception is Recessive q2 - 0.440 q - 0.663 S I 0.159 S I 0.399 So I 1.000 2 1 For Nonsour x Nonsour Mating; 22 Confidence in rejecting hypothesis I 0.87148 For Nonsour x Sour Matingg 21 Confidence in rejecting hypothesis is I 0.11134 For Sour x Sour Hatings 20 I -9.368 F(z) I Z 0 o I 0 Confidence in rejecting hypothesis 2 1.0 I 1.52 F(2) I 0.9357 c = 0.1285 I 0.14 F(2) I 0.5557 0 I 0.8887 Combined Evidence - Conversion to x2 to give probability that Antidesma I Sour perceptions conform to Recessive hypothesis: x: - 18.3311. p < 0.01 1215 Table 32.--Flmily Studies: Antidesma 11 Bitter-Nonbitter Perceptions. Offspring Proportion Proportion Hating Types No. Total of Bitter of Bitter Nonbitter Bitter Observed Expected Nonbitter x Nonbitter 64 100 27 127 0.2126 0.1669 Nonbitter x Bitter 43 46 37 83 0.4458 0.3414 Bitter x Bitter 5 7 2 9 0.2222 1.0000 Total 112 153 66 219 Test of Randomness of Bitter-Nonbitter Perceptions of Offspring x: - 13.245, p < 0.05 Test of Genetic Hypothesis: Bitter Perception is Recessive q2 - 0.2686 q - 0.5183 52 I 0.1669 51 I 0.3414 50 I 1.0000 For Nonbitter x Nonbitter Hatipgs 22 I 1.26 F(z) I 0.89617 0 I 0.2166 Confidence in rejecting hypothesis I 0.7834 For Nonbitter x Bitter Hatings 21 I 1.91 F(z) I 0.97193 a I 0.0561 Confidence in rejecting hypothesis I 0.9439 For Bitter x Bitter Matings 7 7 z0 - -5.6118 F(z) - 2.87 x 10' 6 . 5.74 x 10' Confidence in rejecting hypothesis I 0.999995 Combined Evidence - Conversion to x2 to give probability that Antidesma 11 Bitter perceptions conform to Recessive hypothesis: x: - 20.821, p < 0.005 113(5 Table 33.--Family Studies: Antidesma 11 Sweet-Nonsweet Perceptions. Offerspring Preportion Proportion Hating Types No. Total of Sweet of Sweet Nonsweet Sweet Observed Expected Nonsweet x Nonsweet 83 135 20 155 0.1290 0.0842 Nonsweet x Sweet 27 39 22 61 0.3607 0.2901 Sweet x Sweet 2 2 l 3 0.3333 1.0000 Total 112 176 43 219 Test of Randomness of Sweet-Nonsweet Perceptions of Offersing 2 x2 - 15.2375, p < 0.05 Test of Genetic Hypothesis: Sweet Perception is Recessive q2 - 0.167 q - 0.4087 52 I 0.0842 51 For Nonsweet x Nonsweet Mating§ I 0.2901 50 I 1.0000 Z I 1.665 F(z) I 0.95254 6 I 0.0949 2 Confidence in rejecting hypothesis I 0.9051 For Nonsweet x Sweet Matings 21 I 1.148 F(z) I 0.87493 aI 0.2501 Confidence in rejecting hypothesis I 0.7499 For Sweet x Sweet Mating; z - -2.449 F(z) - 0.00714 6 - 0.0143 0 Confidence in rejecting hypothesis I 0.9857 Combined Evidence - Conversion to x2 to give probability that Antidesma 11 Sweet Perception conform to Recessive hypothesis: x: - 16.0177, p < 0.02 137 Table 34.--Fami1y Studies: Antidesma II Sour-Nonsour Perceptions. Offspring Proportion Preportion Mating Types No. Total Of Sour Of Sour Nonsour Sour Observed Expected Nonsour x Nonsour 13 12 13 25 0.520 0.1822 Nonsour x Sour 60 71 51 122 0.4180 0.4269 Sour x Sour 39 28 44 72 0.6111 1.0000 Total 112 111 108 219 Test of Randomness of Sour-Nonsour Perceptions of Offspring x: = 6.8337, p < 0.05 Text of Genetic Hypothesis: Sour is Recessive 2 q = 0.555 q = 0.745 52 = 0.1822 81 = 0.4269 80 = 1.0000 For Nonsour x Nonsour Matings 22 = 3.381 F(z) = 0.99964 0 = 0.00072 Confidence in rejecting hypothesis = 0.99928 For Nonsour x Sour Mating; Z1 = -0.1991 F(z) = 0.42074 0 = 0.84148 Confidence in rejecting hypothesis = 0.15852 Four Sour x Sour Matingg 20 = -6.765 F(z) = 10'10 a = 20‘10 : 0 Confidence in rejecting hypothesis I 1.0 138 nontasting. Further inspection of these pedigrees however, shows that the results obtained fer Antidesma taste perceptions are ambiguous, in that there are some instances which provide evidence in support of the hypothesis that certain basic taste qualities may be consistent with a recessive inheritance mode. In other instances there is evidence to the contrary. For example, in family #59, the left side of the pedi- gree is consistent with the hypothesis that the sweet perception of Antidesma I may be recessive. Furthermore, in families #62 and #63, there is evidence to suggest that sour perceptions of Antidesma I may be inherited as a recessive trait. It may be recalled however from analyses of the two generation family data that these hypotheses were untenable and that the only hypothesis which had some measure of sup- port was that suggesting recessivity for the bitter perception of Antidesma I. This latter hypothesis however, is not supported by data from family #63 (right side of pedigree) in which a bitter x bitter mating produced a nonbitter offspring. With respect to taste perceptions of Antidesma II, it may also be recalled that the dominant-recessive hypotheses for each of the taste responses were rejected with a high degree of confidence based on the two generation data. While the three-generation data may in some instances support this previous premise as in family #59 (right side of pedigree) where an Antidesma II bitter x bitter mating produced a nonbitter offspring, data from family #62 may suggest otherwise (Antidesma II bitter x bitter mating produced a bitter offspring). Other three—generation data provides no further elucidation of possible dominant-recessive mode of inheritance for specific taste perceptions of Antidesma I or Antidesma II. 139 Analysis of Taste Perceptions of Twins As previously reported, this study included taste perceptions of twelve pairs of twins. Although zygosity was not confirmed by direct serological determinations, according to statements made by parents of these individuals, there were five monozygous and seven dizygous twin pairs. Additionally, as was the case with the three-generation family data, the limited number of twins sampled precluded detailed analyses. Taste perceptions of the three concentrations of PTC and the two Antidesma solutions as reported by twins are recorded in Table 35. As shown, fer each of the PTC concentrations, monozygous (MZ) as well as dizygous (DZ) twins most often judged these solutions as bitter. For MZ twins, the only other taste quality reported for the low and medium PTC concentrations was tasteless and none of them fbund the high concentration tasteless. A substantial number of DZ twins recorded the tasteless perception for all three concentrations of PTC and two of these twins reported salty and sour perceptions. It is interesting to note that in the latter case, 3 DZ twin perceived the low and high PTC concentration as sour but judged the medium concentration as bitter. With regards to perceptions of the Antidesma solutions, none of the twins judged either of these as tasteless or salty. For Antidesma 1, M2 twins most often perceived this as sweet or sour. Similarly, DZ twins often responded sweet or sour to this solution but an appreci- able number (four of feurteen) also feund it bitter. Perceptions of Antidesma II of MZ twins were either sour or bitter while 02 twins reported these as well as sweet perceptions. When the above percep- tions of PTC and Antidesma for M2 and DZ twins were compared, no sig- nificant differences were observed for these groups (p > 0.05). 140 NN.N A N .NN.N Nx NNN NENNNNN=< NN.N A N .NN.N u Nx “HN\NN NN.NNN NNN: NNN N N NN.N A N .NN.N 1 MN ”N NNNNNNNNN NN.N A N .NN.N u MN "HN\Ns NN.NNN No: NNN NN.N A N .NN.N 1 mx "HN\NN NN.NNN zoN NNN mcwzk maowNNNa Noszmm< u No mcwzk msowxucoz NeESmm< 0 N2« NH oH NH oH NH oH NH 0H NH oH Hmuoh N N N N N N N N N N NNNNN N N N H m oH w m N N Houuwm N o o m o o o o o o u003m w m N N H o o o H o usom o o o o N c m H m m mmoHopmmH N9 N2 N0 N2 No N2 N9 N2 Ne NZN HH mEmoNNuc< H «ENNNNNN< .ocou :wN: .ocou .Noz .ocou so; mammnNuc< DEN .ch39 mo NENNNNNN< Nam UHN mo mcowuaoonoN pummeuu.mm oHan 141 To possibly elucidate the relative role of genetic factors in determining taste perceptions, concordance rates for M2 and DZ twins were computed and compared. These rates fer each of the PTC solutions are reported in Table 36a. As indicated, the concordance frequencies have been calculated for the overall (actual) PTC perceptions as well as those for bitter-nonbitter and taster-nontaster perceptions. As can be observed, concordance rates for M2 twins are the same for each of the three types of computations for a given PTC concentration and with the exception of the taster-nontaster classification, were higher than those of DZ twins. When this latter dichotomy was employed, four of the five MZ twins reported identical perceptions for the low and medium PTC con- centrations for concordance rates of 0.8. This may be contrasted with a concordance rate of 0.857 for DZ twins fer these same solutions. It should be noted however, that this value, unlike that of the MZ twins was derived not from identical taste perceptions in these twins but from their classification as nontasters or tasters (regardless of taste per- ception recorded). A similar situation was observed fer the high PTC concentration in which the concordance rate fer both M2 and DZ twins was 1.0. While each of the M2 twin pairs perceived this solution as bitter, in two separate instances one member of a DZ pair responded bitter while the other member of the pair judged the solution as sour or salty. Since by definition all of these individuals were considered tasters however, the concordance rate of 1.0 for the M2 and DZ twins may not be strictly comparable. For purposes of testing equivalence of concordance rates for taste perceptions of monozygous and dizygous twin pairs, Fisher's Exact Probability Test was employed (fer rationale and explanation of this 142 N.N u N NNN.N u N NNN.N u N "NNN: NNN "NNNNNNNNNNN NNN.N u N NNN.N u N NNN.N u N ANN: NNN poaxm NNN.N u N NNN.N u N NNN.N n N "zoN NNN N.NNNNNN N.N N.N NNN.N N.N NNN.N N.N NNNNzN NNN NNN.N N.N NNN.N N.N NNN.N N.N NNN2N NNN NNN.N N.N NNN.N N.N NNN.N N.N HzoNN NNN NN N2 NN N: NN N2 mcoNumoouoN NeummucoZuuoummN mcoNuaooNoN NvuuflncoZINouuNm mcoflumooHoN NHmauoo .moumm oocmcnoocoo NNNN39 mo mcoflumooNoN cummh UHNnu.mom oHnmh 143 procedure, see Appendix). When this test was applied to these con- cordance rates for overall, bitter-nonbitter and taster-nontaster per- ceptions for each concentration of PTC as reported in Table 36a, dif- ferences between the twin types were not significant (p > 0.05). Concordance rates fer Antidesma taste perceptions of twins are presented in Table 36b. As can be observed, these rates have been com- puted for the actual perceptions reported as well as for bitter- nonbitter perceptions. With respect to the overall taste perceptions of Antidesma I and II, concordance rates fer monozygous twins were some- what higher than those of dizygous twins. For the bitter—nonbitter per- ceptions, MZ twins were also concordant more often than the DZ twins for Antidesma I but not fer Antidesma II. Comparisons of the concordance rates for each of the Antidesma solutions however, revealed no signifi- cant differences between these rates for the two twin groups (p > 0.05). 144 NNN.N u N ANN NENNNNNNN NNN.N u N NNN NENNNNNN< ”NNNNNNNNNNN womxm NNN N u N ”N NNNNNNNNN NNN.N u N ”N NENNNNNNN N.NNNNNN NHN.o o.o HNm.o o.o HH mamovfluc< mNN.o m.o mNH.o N.o H mamopwuc< No N: No N: mcowuaoohoN Houuwn:OZuNouuNm NcofiummonoN HHm3uoo .mmpmm mocmpuoocou chNzk mo NNONuNmoNoN mumms mamovfluc 51 years) recorded less incor- rect responses than those of younger ages, particularly when compared to ages 18-30 years (see Tables 6a and 6b). The apparent disparity in these results may be due to the use of more concentrated solutions which provided stimuli so far in excess of threshold levels so as to obscure any age effects which might have been noted. Investigations which have examined basic taste quality sensi- tivity with respect to sex have produced conflicting results. A number of these studies have reported the lack of apparent sex differences in taste sensitivity (Aubek, 1959; C°°PBT.EE.El-: 1959; Krut epugl., 1961). In contrast, other studies have suggested that females have greater taste acuity than males. Pangborn (1959) reported that in general, females have lower taste thresholds than males. Tilgner and Barylko- Pikielna (1959) feund women to have a higher sensitivity than men fer sweet and salty but less for sour and no difference between the sexes fer bitterness. Studies by Meiselman and Dzendolet (1967) however, suggest that more males than females consistently confuse the sour and bitter taste qualities. Their work further suggests that except for the identification of sweet, males are less sensitive tasters in that they are more likely to misjudge standard control solutions when com- pared to females. Data from the present study is consistent with this latter finding (see Tables 9a and 9b) although these apparent sex dif- ferences in taste perception were not feund to be statistically sig- nificant. 149 To determine whether a subject's membership in a specific race or ethnic group may contribute to differences in overall taste percep- tions, an analysis of perceptions of controls by race was performed. There are apparently no previously published studies of this type with which to make comparisons. Hewever, as indicated earlier, data obtained in this study have revealed no significant racial taste per- ceptional differences in responses to the control solutions fer spe- cific taste qualities. Thus individuals of different racial groups are just as likely to correctly or incorrectly identify the basic taste qualities (Tables 12a and 12b). Despite the widespread belief that smoking decreases overall taste sensitivity, the experimental evidence is surprisingly inconclu- sive and/or discordant. Krut.ep'el. (1961) have suggested that smokers are less sensitive to bitter, based on their finding of a significantly higher mean taste threshold of these subjects for solutions of quinine hydrochloride while the mean thresholds fer control solutions for sweet, sour or salty were similar in smokers and nonsmokers. A similar insensi- tivity to bitter particularly in heavy smokers was also observed by Fischer et‘gl. (1963). Furthermore, Peterson et_§l. (1968), in an extended study of smokers versus ex-smokers reported a significant decrease in taste thresholds (increased sensitivity) among ex—smokers after one month when compared to those who continued to smoke. In con- trast to these findings, Cooper 33 21. (1959) observed no differences between smokers and nonsmokers in ability to detect any of the four primary tastes. McBurney and Moskat (1975) when measuring both ,1 detection and recognition thresholds of several compounds in smokers and nonsmokers found no consistent differences in either measure 150 between the two groups. While not precisely comparable with the above studies, taste acuity in smokers versus nonsmokers as assessed in the present study by appropriate responses to the control solutions is in agreement with these latter studies in that no significant associations of taste perception differences and smoking status were observed (Tables 15a and 15b). The effects of hunger on taste sensitivity are uncertain. Yensen (1959) reported a significant decrease in sensitivity fer about one hour after a meal followed by an increase in three or four hours. Similar findings have been suggested by some workers (Gusev, 1940) but have not been confirmed by others. Meyer (1952) for example, found no change in sensitivity to taste up to thirty-six hours of fasting. In the present investigation the comparison of taste responses to controls with time of last f00d eaten produced uniform results throughout each elapsed time category. This supports previous observations of a lack of change in general taste sensitivity with time since ingestion of last f0od (Tables 18a and 18b). Taste Perceptions of'PTC Although data derived from numerous population studies indicate that the majority of human subjects perceived PTC as bitter or tasteless, other taste sensations fer this compound have also been noted. Several researchers have feund PTC perceptions of other taste qualities in addi- tion to nontaste quality descriptions (e.g., camphory, sulfury). The reported incidence fer sour PTC perceptions has been 2.3-5.4 percent, fer sweet perceptions, 2.1-8.9 percent and fer salty perceptions, 3.5- 4.8 percent (Blakeslee and Fox, 1932; Blakeslee, 1935; Skude, 1959; 151 Skude, 1960; Richter and Clisby, 1941; Harris and Kalmus, 1949; Amerine et 31., 1965). Corresponding values of these perceptions for the three PTC concentrations used in the present study as recorded in Table 4a, were 3.8-4.9 percent (sour), 0.09-0.3 percent (sweet), and 0.7-0.8 per- cent (salty). It will be noted that while the sour perceptions of PTC obtained in the present study compare favorably with those reported in previous investigations, the values obtained for sweet and salty PTC perceptions are lower than those previously reported. This dis- crepancy may be due to differences in testing procedures employed. For example, most of the previous studies included both lower and higher concentrations than those of the present study. It has been noted by some workers, that the incidence of "abberrant" PTC tasting (sensations other than bitter) increases at lower (subliminal) concentrations (Richter and Clisby, 1941; Skude, 1960; Rychokou and Borodina, 1973). This may also be true of instances where concentrations used are so high that some individuals who are in fact considered "nontasters," based on their perceptions of PTC solutions above the population anti- mode (81.25 mg/l), may have described taste sensations other than the bitter taste usually perceived. The relationship between age and PTC taste sensitivity has been studied extensively but still remains uncertain due to the lack of agreement of published studies. Studies by Harris and Kalmus (1949) and Barnicot (1950) have suggested a deterioration in PTC taste sensi- tivity with age as evidenced by their finding of an increase in threshold perceptions of about one dilution step for each twenty years of age (e.g., 20.31 mg/l versus 40.63 mg/l) up to age fifty. Giles e; .31. (1968) and Ghosh (1973) have also reported marked fluctuations in 152 taster-nontaster frequencies with age but with no consistent age trends. Although several studies have noted some decreased PTC sensitivity with age, most have failed to confirm any significant age effects for threshold levels or PTC tasting status (Mohr, 1951; Paolucci et 31., 1971; Alsbirk and Alsbirk, 1972; Bonne et 31., 1972; Sriram et 31., 1975; Ingley 23.31., 1976; Ibraimov et_§1,, 1977). Results from the present study are in agreement with these latter findings in that no significant age differences in taste perceptions were found fer any of the three PTC concentrations employed (Tables 8a and 8b). Investigations of the relationship of sex and PTC tasting have produced fairly consistent results. As noted previously, females have been found to be more sensitive tasters in that they can detect PTC in higher dilutions than males. A few studies have reported this differ- ence to be significant (Falconer, 1946; Montenegro, 1964; Giles 25.31., 1968; Scott-Emuakpor.et.§l., 1975). Most reports however, have noted only nonsignificant threshold differences between the sexes (Hartman, 1939; Barnicot, 1950; Mohr, 1951; Soltan and Bracken, 1958; Bonne et .31., 1972; Glaser, 1972; Than-Than-Sint and Mya-Tu, 1974; Ingley et .31., 1976; Ibraimov et-gl., 1977; Tandon and Pandey, 1978). The present study is in general agreement with a majority of these reports since no significant associations of overall PTC perceptions with sex were feund (Table 11a). Furthermore, despite the slightly greater fre- quency of female tasters, no significant differences in the sex-related proportions of tasters and nontasters were observed fer each of the three PTC concentrations used (Table 11b). However, as also noted in Table 11b, when PTC perceptions were classified as bitter or nonbitter, significant sex differences were observed but only for the low 153 concentrations of PTC (20.31 mg/l). These findings along with others previously discussed suggest the need for further investigations of the relationship of sex and taste perceptions in general as well as sex, PTC taste sensitivity and bitter cognition in particular. As indicated earlier, numerous population studies have noted considerable racial or ethnic differences fer PTC perceptions. Most of these have indicated that Negroid, Mongoloid and American Indian popu- lations are generally characterized by a nontaster frequency of less than 20 percent, while this figure in Caucasian populations is usually 25-35 percent (Lee, 1934; Cohen and Ogden, 1949; Barnicot, 1950; Lugg, 1966, 1968, 1970; Mohr, 1951; Monn, 1969; Sunderland, 1966; Sunderland and Rosa, 1975; Barnicot, 1950; Barnicot and Woodburn, 1975; Bhalia, 1972; Scott—Emuakpor.et.gl., 1975; Frisancho e: 21., 1977; Bhalia, 1972; Allison and Blumberg, 1959; Giles etngl., 1968; Srivastava, 1974; Erikson .EE.El" 1970; Jenkins, 1965; Mitchell ep‘gl., 1977; Corcos and Scar- borough, 1978). That there is much variation in these reported race/ ethnic group frequencies can be seen from data compiled in Table 37 and probably reflects the diversity of sampling techniques used, as well as sample size of the populations tested. Of the six race/ethnic groups sampled in the present study, only two groups, the White/Caucasians and Black/Afro-Americans had sufficient numbers represented to facilitate comparisons. Using the high concen- tration of PTC (81.25 mg/l) to distinguish tasters from nontasters, the finding of 21.5 percent nontasters in the White/Caucasian group is in good agreement with previously reported data and the value of 17.7 per- cent nontasters in the Black population is also well within limits of values derived from earlier studies. It will also be noted that the 154 N.mN Hmocwsu 3mzv moNENNN o.N mocflwflpon< N.NN NNN>NNN2N NNNNNNNN=< N.NN NNNNN: o.NN mcmwumxwm o.wH NmNmHmzv mouNNmoz N.NN NNNNNNN NNNN< N.NN-N.N NNNNNNNNNN N.NN NNNNNNN NNNNN N.NN NNNNNN NNNNNNNNN N.NN-N.NN NNNNNNN NNNN< N.NN-N.N NNNNNN .N.= N.mm NDNNN: N.mH-m.N mxomHm :moNnm< a.mN mNcmNH N.NN nmficmmm N.NN NNNNNN N.NN NNNN: N.NN-N.NN NNNNNNN NNNENN N.NN NNNNNNNNNN m.mN HNNNmmmv mENHmsz o.Nm HocNumoHan mmuNNNEom N.NN-N.NN HNNNNNN NENNNNz N.NN NemNeosN< N.N -N.N N.NNE< .NNNN.zN NNNNNNN N.NN NNNNNN< .zN :mNzNN N.N -N.N N.NNE< .zN NcaNNNN N.NN HNNNNNNNNNN NNN26N a.mNum.mN NoeNxmm N.NN NNNan NNNHNNNNN o.m mcmonox N.Hm mcmflmmsm N.NH NcmNm< HmNucoo N.NNum.oH :NNNNNN N.NN NNNENNN N.NN NNNNsz o.m uo.N ommcmmmw m.omuo.NH mcmwmozhoz o.HHuw.H mammoENom w.Hm :mficm: N.N NNNNNNN N.NN-N.NN NNNNNN .N.= N.NN-N.NN NNNNNNNN N.NN-N.NN NNNNNNN NNVNocmzcoNN :oNumHSNON NNN Nocoacoum :oNpmHSNON .Hmouusom HmNo>om sonny mpoummucoz QNN mo moNocosuoNN NONumHsmoNnu.Nm oHnmN 155 difference in the nontaster frequencies obtained for these two groups is statistically significant (see Table 14b). In this study an attempt was made to assess the effects of smoking on PTC perceptions by comparing responses of smokers and non- smokers. Although no significant association of smoking status and overall perceptions for each PTC concentration was observed, the dis- covery that smokers consistently were more likely to find each of these solutions tasteless suggests that smoking may produce some effect (Table 17a). The additional finding that differences in the absolute frequencies of tasters and nontasters in these two groups were indeed significant for the low PTC concentration, although not for medium and high concentrations, strongly suggests that smokers may have reduced taste sensitivity and thus have higher PTC thresholds. These results are at variance with some of the previous smoking and PTC studies but are in agreement with others. Falconer (1946) reported no apparent threshold differences between smokers and nonsmokers and Salmon and Blakeslee (1935) found no strong correlations between use of tobacco and PTC sensitivity. Krut gt g1. (1961) reported a higher but not significant mean threshold in smokers and Fischer et 31. (1963) found fewer smokers (especially those who smoked at least fifteen cigarettes per day) among sensitive tasters with the lowest thresholds but the difference was not statistically significant. Hall and Blakeslee (1945) however, concluded that smoking reduces acuity to PTC and Leguebe (1969) as well as Thomas and Cohen (1960) found a signifi- cant association between high PTC thresholds and smoking. In spite of the lack of agreement of the effects of smoking on PTC thresholds, it is interesting to note that all of these studies, including the present 156 one concur that the distribution of tasters and nontasters is similar in smokers and nonsmokers (as assessed by perceptions of 81.25 mg/l), however, in light of all of the above findings, it is also clear that the relationship of smoking and PTC threshold sensitivity deserves further investigation. Assessments of the effect of time since last f00d eaten on PTC perceptions made in this study produced unexpected and interesting results. The frequencies of overall taste responses as well as propor- tions of tasters and nontasters in each elapsed time category were similar for each PTC concentration. However, a significant difference was observed for the bitter versus nonbitter responses for the highest concentration (Tables 20a and 20b). Additional intriguing results were the significantly lower proportion of tasters than nontasters for the high concentration and less bitter than nonbitter responders for the medium and high concentrations when these perceptions were compared, at elapsed times of less than versus greater than one hour (Table 20c). These data suggest a decreased sensitivity to PTC in general and its perception as bitter in particular, during the first hour after the ingestion of food. These findings, reminiscent of those suggested from hunger studies by Yensen (1959) and Gusev (1940) which were men- tioned earlier, have not been previously reported for PTC and thus should be worthy of extended study. Taste Perceptions of Antidesma The present study has confirmed the diversity of taste responses of Antidesma as initially reported by Henkin and Gillis (1977). These differences in taste responses were found both for Antidesma aqueous 157 extracts (Ad I) and for liquified Antidesma macerated material (Ad 11). While perceptions of this latter solution cannot be strictly compared with the Henkin and Gillis report, it was included in the taste sampling because preliminary results from a small pilot study suggested a less than expected proportion of bitter responders (< 5 percent) for the aqueous extract while an appreciable number of individuals (> 10 percent) who sampled the actual fruit were bitter responders. It was thought that perhaps the major factor(s) responsible for eliciting the bitter response might reside in parts of the fruit other than the aqueous extract. When overall taste responses to both Antidesma I and II were examined, it was observed that the frequency of misclassification of control solutions had no apparent effect on the perceptions of the Antidesma solutions (see Tables 3a and 3b) thus suggesting that most individuals sampled were able to recognize the basic taste qualities. There was considerable diversity of taste responses for each Antidesma solution ranging from 0.6 percent salty to 50.9 percent sweet for Antidesma I and 0.0 percent tasteless to 45.9 percent sour for Antidesma 11. Other perceptions of Ad I were tasteless-~0.8 percent, sour—-36.0 percent and bitter--ll.7 percent and for Ad 11, 25.4 percent sweet, 0.5 percent salty and 28.2 percent bitter. Subjects also recorded a greater mean intensity for Ad 11 than f0r Ad I (3.098 versus 2.493). When the taste perceptions of Ad I and II were compared, highly significant differences were found for overall responses as well as for dichotomous categories (e.g., bitter-nonbitter, sweet-nonsweet, etc.) of the major perceptions of these solutions (see Tables 24a, 24b and 158 24c), thereby suggesting major differences in composition of the two Antidesma preparations. Table 38 compares Antidesma perceptions from the present study and those reported by Henkin and Gillis. It can be observed that while the absolute proportions of responders and nonresponders to aque- ous extracts were not identical in both studies, the differences were not significant. Conversely, differences in responses for the Antidesma extract of Henkin and Gillis and those of the Antidesma II (macerated material) of the present study were highly significant (p < 0.01). The favorable comparison of responses to the aqueous extracts in both investigations suggests that these two solutions are quite similar and that they are dissimilar to the macerated material (Ad 11). These data additionally show that components of the macerated material also strongly elicit bitter perceptions as evidenced by the even greater frequency of bitter responders to Ad 11. Whether this represents a con- centration effect and/or additional bitter evoking factors is presently unclear but may be elucidated by further research. As noted earlier, overall perceptions of Antidesma solutions in this study were not strongly associated with age. However, when the bitter-nonbitter (responders-nonresponders) classification was employed, significant differences were observed among the various age groupings, although no definitive age trends could be discerned (Tables 7a and 7b). These findings do not concur with those of Henkin and Gillis who reported no relationship in responsiveness to Antidesma with age. Failure of these workers to note any age effects may have been due to their smaller sample size (170 versus 1,438 subjects) 1559 momcoamou wouuwncoz u muovconNohcoz noncogmou wouuHm u muovcommomw NN.N v N .NN.NN u Mx NNN N< NNN: N-= NN.N A N .NNN.N . “x "N u< NNN: N-= N.NN NNNN N.NN NNNN N.NN NNN NNNNNNNNNNNNz N.NN NNN N.NN NNN N.NN NN NuNNcoNNoN N .62 N .62 N .oz 6 H9935 oudkoual HUME uxu NN N N N NN N< NNNNNNNNN N N< NNNNNNNNN Nuaum acomouN Huuauuxo Naoomcav HN-:N NNNNNN N NNNNN: N.Nuovconoucoz New Nuovconom mamovNuc< mo comNuunsou-u.wn oHnah 160 and/or the greater mean age of subjects (43 years versus 21.9 years). It is probable that this latter factor may be the more contributory one since in the present study, greater deviations in frequencies of responders to Antidesma extract were found for younger age groupings. The lack of correlations between taste perceptions of Antidesma aqueous extracts and sex of respondent as reported by Henkin and Gillis has been supported by results from the present study. No significant sex differences were observed for overall perceptions for both Antidesma I and 11 nor for bitter-nonbitter responses for Antidesma I (Tables 10a and 10b). It is interesting to note however, that significant male-female differences were observed in bitter-nonbitter responses to the Ad 11 macerated material (not included in Henkin- Gillis study). The greater frequency fer female bitter responders as well as the general tendency for all subjects to more often judge this solution as bitter when compared to Ad I is suggestive of the increased sensitivity to bitter in females as discussed earlier. In the current investigation, racial variations in Antidesma taste responses were observed (Tables 13a and 13b). These differences were found to be highly significant especially when bitter and nonbitter responses of Black and White subjects were compared. Bitter responders among Blacks were two to three times more frequent than among whites. These results are in conflict with those of Henkin and Gillis who found no correlation between race or national origin and Antidesma responses. That these researchers were unable to detect race or ethnic diversity for Antidesma perceptions which is so strikingly evident from the present study was probably due to the relative racial homogeneity of their subjects (160 Whites, 8 Blacks and 2 Orientals). 161 Although not explored in the previous Antidesma investigation, the present work revealed no relationship between smoking status and Antidesma responses (Tables 16a and 16b). Similarly, the time of last food eaten apparently had no appreciable effect (Tables 19a and 19b). Antidesma Perceptions and PTC Responses Based on their research, a major conclusion reached by Henkin and Gillis involved the specific association of taste perceptions of Antidesma and PTC since no single individual in their study was a responder (had bitter perceptions) to both of these substances. To examine the validity of this conclusion, taste perceptions of Antidesma I and Antidesma II were compared to responses to three concentrations of PTC. (Different PTC concentrations were used since the exact con- centration employed by Henkin and Gillis was not stated in the original report.) No significant differences were observed for any of the spe- cific perceptions of Antidesma I (comparable to extract used in the original study) with respect to any specific taste response to each of the PTC concentrations. Furthermore, the proportions of PTC tasters and nontasters were randomly distributed with respect to specific percep- tions of Ad I (Tables 21a, b, and c; 22a, b, and c; and 23a, b, and c). In addition, 91 subjects found both Ad I and the low PTC concentration bitter, 102 responded bitter to Ad I and medium PTC and 111 individuals judged Ad I and PTC high as bitter (Tables 21c, 22c, and 23c). These values represented 6.3 percent, 7.1 percent and 7.7 percent respectively of the total population sampled and therefore, are not in agreement with the original report. When Antidesma II (macerated material) and PTC perceptions were compared, discordant results were obtained. While 162 comparisons of Ad 11 perceptions with respect to overall responses to each PTC concentration were similar, these Ad 11 perceptions were sig- nificantly different when compared to PTC bitter-nonbitter responses at each concentration (Tables 21b, 22b, and 23b). A similar significant difference was seen when Ad 11 bitter-nonbitter groupings were also com- pared to the bitter-nonbitter dichotomy for each concentration of PTC (Tables 21c, 22c, and 23c). While results obtained utilizing the macerated material cannot be directly related to the Henkin and Gillis data, because the significant differences observed were mainly due to the less than expected frequencies of individuals who judged both Ad 11 and PTC as bitter, this does suggest that some relationship of bitter cognition for these two substances may exist although not as strict as proposed by the original study. The above findings pose an interesting problem. It is not evi- dent why the results obtained from the macerated material of Ad 11 sug- gest a possible, although limited relationship to that of the Henkin- Gillis data while those obtained with the aqueous extract (Ad I) fail to provide evidence in support of their data. Possible explanations may involve concentration differences for both the Antidesma and PTC solutions and/or variations in sampling techniques as alluded to earlier. In spite of potential reasons for discrepancies with the previous report, overall results from the present study clearly do not support the mutual exclusivity of bitter perceptions of Antidesma and PTC as observed by Henkin and Gillis. 163 Family Studies of Taste Perceptions The majority of taste perception data in families was derived from analysis of the 112 two generation families sampled. These results showed that offspring had slightly greater but insignificant misclassification rates for control solutions when compared to parents. Offspring on the average also recorded higher intensities for each of the controls than their parents (Tables 25a and 25b). This may be sug- gestive of some possible age effect. Additional variation between these two groups was seen in their overall taste responses to Antidesma I where significant differences between parents and offspring were observed. Whether these findings are due to judgemental differences or real sensory variations is unclear. However, no such variation was found for progeny versus parental perceptions for Antidesma 11 nor for any of the PTC concentrations (Tables 26a, 26b, 27a, and 27b). Genetic Analysis of PTC and Antidesma Taste Perceptions Statistical analysis by Chi-square of the offspring resulting from the various mating types with respect to PTC taster-nontaster pheno- types revealed highly significant differences between observed and expected progeny thus implying that these results were not likely to be due to chance alone. When these data were subsequently analyzed for their concordance with the genetic hypothesis, that PTC tasting is dominant and nontasting is recessive, it was found that they were in excellent agreement (p > 0.95) with this well established theory (Table 28). This close agreement was further substantiated by the three generation PTC family data in which no exceptions to this hypoth- esis were observed (Fig. 5, families 59, 62 and 63). 164 When similar statistical treatment was performed for the dichotomous classifications (e.g., bitter versus nonbitter, sour versus nonsour, etc.) of the two Antidesma solutions, with the assumption that the basic taste perceptions were recessive, divergent findings were observed. With respect to Antidesma 1, tests fer randomness for observed and expected frequencies of offspring resulting from the various matings disclosed that only the bitter—nonbitter perceptions appeared to be nonrandom, while the sweet-nonsweet and sour-nonsour perceptions could be accounted for by chance (p > 0.05) (see Tables 29, 30 and 31). Upon subsequent testing of the proposed genetic hypothesis for these perceptions, it was found that there was support for a dominant-recessive mode of inheritance for the bitter versus nonbitter perceptions (0.5 < p > 0.1) while this hypothesis was rejected with a high degree of confidence for the other taste perceptions (p < 0.05). Somewhat different results were obtained from analysis of the Antidesma II two-generation family data (Tables 32, 33 and 34). Although the major taste perceptions of this solution appeared unlikely due to chance (p < 0.05), the overall evidence strongly supported rejection of a dominant-recessive pattern of inheritance since the probability that these data conformed to this proposed hypothesis were all less than 0.02. Taste perception data for Antidesma I and II from the three generation families were ambiguous in that there was support for pre- viously rejected recessive hypotheses for sweet and sour perceptions of Antidesma I but lack of support for possible recessive nature of the bitter perception accepted earlier. Likewise, Antidesma II taste perception patterns observed in these families in one instance 165 supported the absence of dominant-recessive inheritance for the bitter perception while in another case the reverse was seen. Due to the small numbers of these families and the absence of results for several first generation members, definitive conclusions were unwarranted. From the foregoing discussion, it may be noted that based on evidence from the majority of the families studied, a dominant-recessive hypothesis was primarily supported only in the case of bitter-nonbitter responses for Antidesma I. It is not clear why these same responses obtained for Antidesma 11 did not produce similar results. It may be that additional bitter evoking factors in this second solution were unrelated to those in Antidesma I. On the other hand, if these bitter response-causing agents are similar in both solutions, it may be that Antidesma 11 contained a.much greater concentration of these, producing results which obscured genetic tendencies in favor of the tested hypoth- eses. If such is the case, these results would be somewhat analogous to those obtained in some PTC studies when tasters and nontasters are iden- tified by use of concentrations which are much higher than the popula- tion antimode (> 81.25 mg/l). The additional probability of nonrandom- ness observed for each major perception of Antidesma II was also enig- matic. Whether this represents specific intervening environmental or other genetic factors was not obvious from this investigation. Taste Perceptions of Twins Analysis of taste perception data for the twelve pairs of twins in this study produced uniform results. No significant differences were observed between the monozygous and dizygous twins for each of the PTC concentrations and f0r the two Antidesma preparations (Table 35). 166 Similarly, when concordance rates for these two types of twins were examined, no significant differences for overall PTC and Antidesma responses, for bitter versus nonbitter perceptions nor for PTC taster- nontaster frequencies (Tables 36a and 36b). These data suggest a lack of strong genetic influence on taste perceptions of these substances and appear to be at variance with a previous report. Although there are no prior studies of Antidesma perceptions in twins, Martin (1975) from an investigation of PTC tasting in twenty-eight MZ and eighteen 02 twin pairs reported a significant variance in concordance thresholds between the two twin groups. These findings may not be strictly com- parable with those of the present study since in the previous investi- gation, thresholds were assessed by use of fourteen different concen- trations of PTC, there was serological determination of twin zygosity and a much larger population of twins was sampled. Definitive conclu- sions from the present study may therefore be severely limited. It is interesting to note however, that calculations of the probabilities of similarity of concordance rates for MZ versus DZ twins produced higher probabilities (p = 0.318-l.0) for the PTC data than for the Antidesma results (p = 0.221-0.S30). Since differences in PTC perception are known to be genetically determined, the greater similarity between M2 and DZ twin concordance rates for PTC than for Antidesma may suggest that genetic influences on Antidesma perceptions should not be ruled out. However, definitive conclusions are unwarranted due to the small sample size of twins studied. 3. Summary- The principal purposes of this study, prompted by the previ- ously reported association between taste perceptions of PTC and 167 extracts from the fruit of Antidesma bunius, were to assess taste responses fer these substances by age, sex and racial groupings, to study associations between PTC and Antidesma taste responses, to determine from family studies if taste perceptions of Antidesma could be accounted for by a simple dominant-recessive genetic hypothesis and to ascertain if additional factors such as smoking status and time of last food eaten have effects on these taste perceptions. The addi- tional use of solutions as controls for the various taste qualities also allowed estimates of the reliability of taste perceptions recorded, as well as quantification of misperceptions of substances typically per- ceived in a certain manner by a majority of human subjects. Towards these ends, taste responses to standard control solutions, three concen- trations of PTC and two preparations of Antidesma were assessed for 1,438 subjects which included unrelated individuals and family group- ings. A summary of the major findings from this study are listed below. 1. Misclassification rates of 1.4 percent to 17.9 percent for control solutions for the taste qualities of sweet, tasteless, salty, sour and bitter were found to be much less than those reported from previous studies, although in general agreement with other reports, the greatest tendency for misclassification occurred in the distinction between bitter and sour. Perceptual errors in controls did not appear to be significantly affected by race, age, sex, smoking status nor elapsed time since last food eaten. Additionally, misidentifications of controls did not appear to produce significant differences in taste responses to PTC and Antidesma. 168 2. The majority of subjects sampled judged each of the PTC solutions as bitter or tasteless as expected, however, other percep- tions ranging from 0.09 percent for sweet to 4.9 percent for sour were also reported. Based on responses to the PTC solution concentration of 81.25 mg/liter, the overall incidence of tasters was 75.8 percent and of nontasters, 24.2 percent. Corresponding bitter-nonbitter responses at this concentration were 70.1 percent and 29.9 percent respectively. 3. Analysis of PTC perceptions by age groupings of subjects who ranged from ages seven to seventy-two showed no significant age effects on overall perceptions, taster-nontaster frequencies nor bitter-nonbitter responses. 4. Comparison of PTC perceptions for the 620 males and 818 females revealed no significant sex differences fer overall perceptions nor taster-nontaster frequencies. The slightly greater proportion of female tasters at each concentration level and the significantly higher frequency of female bitter responders for the low PTC concentration sug- gests that females may have greater taste sensitivity to this substance. 5. The frequencies of nontasters in the 1,213 White/Caucasians and the 198 Black/Afro-Americans were 21.5 percent and 17.7 percent respectively. These differential racial frequencies were found to be statistically significant and were in good agreement with values reported from previous studies. 6. No significant associations of smoking status and overall PTC perceptions were found when responses of 258 smokers and 1,180 nonsmokers were compared, however, smokers were consistently more likely to describe each PTC concentration as tasteless and the 169 proportion of nontasters among smokers was significantly higher than that of nonsmokers for the low PTC concentration. This suggests that smoking may reduce PTC taste acuity. 7. Elapsed time since last food eaten appeared to have a sig- nificant effect on PTC perception especially in the case of bitter responses. This effect seems most pronounced within the first hour after food ingestion and was evidenced by a significantly lower pro- portion of tasters of the high PTC concentration and less bitter responders fer the medium and high concentrations. 8. Comparisons of taste perceptions of the two Antidesma prep- arations used (aqueous extract and liquified macerated material) revealed significant differences in overall responses as well as for dichotomous classifications of the major perceptions of these solutions which is suggestive of inherent compositional differences. 9. No significant effects of age on the overall perceptions of the two Antidesma solutions were observed however significant differ- ences among age groupings were found when perceptions were classified by the bitter-nonbitter dichotomy although no specific age trends could be discerned. 10. When compared to males, a significantly greater proportion of females were bitter responders for Antidesma II (macerated material). No other significant sex differences were observed for either the overall perceptions of both Antidesma 1 and II or bitter-nonbitter responses for Antidesma I (aqueous extract). 11. Highly significant racial differences between Blacks and Whites were found for Antidesma bitter-nonbitter responses. 170 12. No apparent effects of smoking on Antidesma perceptions were evident. Likewise, elapsed time since last food eaten produced no absolute effects. 13. There were no significant associations observed for any specific taste perceptions of Antidesma I with any taste response to each of the PTC concentrations. Frequencies of PTC tasters and non- tasters were also randomly distributed with respect to Antidesma I per- ceptions. Conversely, overall perceptions as well as bitter-nonbitter perceptions of Antidesma 11 showed significant correlations with PTC responses primarily due to the less than expected frequency of individ- uals who judged both Antidesma II and PTC as bitter. There was how- ever, no mutual exclusivity of bitter perceptions for either Antidesma II or I and PTC. 14. Analysis of PTC taster-nontaster progeny frequencies from various mating types showed close agreement with the generally estab- lished dominant-recessive hypothesis. Support for this hypothesis for the Antidesma taste perceptions in families was found only in the case of bitter-nonbitter responses for Antidesma I. 15. Comparisons of twin concordance rates for Antidesma per- ceptions revealed no significant differences between concordance of M2 and DZ twins. The probabilities of similarity of concordance rates for M2 versus 02 twins was higher for the PTC than for the Antidesma results. APPENDICES APPENDIX A DERIVATION 0F SNYDER'S RATIOS APPENDIX A DERIVATION 0F SNYDER'S RATIOS Frequencies of Mating Types and Offspring Offspring Mating Types Frequency TT Tt tt TT TT p4 p4 TT Tt 4p3q 2p3 2p3q TT tt 2p2q2 2qu2 Tt Tt 4p2q2 p2q2 2p2q2 p2q2 Tt tt 4pq3 2pq3 2pq3 tt tt q4 q4 1 so that: _ R 2 - D + R Percent Recessives from Dominant x Recessive Matings (one parent = dominant) = SI: (TT x tt and Tt x tt Matings) Dominant Progeny = 2p2q2 + 2pq3 = ZPQZ 3 Recessive Progeny = 2pq so that: R ‘ 0 + R - 3 3 qu_ - qu _ 9. _ q ‘ 2 2 3 " 2 ‘ + 2 ' 1 + 9 2p q + 499 299 (p + 299) p q Percent Recessives from Dominant x Dominant Matings (both parents = dominant) = $2: (TT x TT, Tt x TT, Tt x Tt Matings) Dominant Progeny = p4 + 2p q 3 + szq2 = p2(L + Zq) Recessive Progency = p q ._ _... _. 2 2 2 - P Q. - ___£L____ [p211 + 2q) + (p2q211 (1 + q)2 171 APPENDIX B COMPARISON OF PTC AND ANTIDESMA RESPONDERS AND NONRESPONDERS (HENKIN AND GILLIS, 1977) om mag mm mnmvcommmncoz ohm muovcommom ohm muovcommoucoz «Emovfiuc< mhmvcommmm «Emmvflu:< m.mw n.v~ hwy xocoSUonm omcoammm :moz anew Hzom mmmv Houuwm A.oz .omcommomu xufiamso mmconmom suave: NH om hwy xuwmcoucu omcommmm cmflcoz Aoofi-ov «N fioofi-mv mm Amman“ .wv xwfimcoucH omcoamom cam: Amn-mv Ne Amw-nv we Amman“ ..mng ow< awe: mmHmEom mm ammfime an mofimemm m .mofiae 0H xom 3; mm 3833 mammkuc< v.~m 0.50 mwv xocoscoum mmcommom :moz Anvv mmofimummp mmHHV youuwm m.oz .mmcommomv xufifimso mmcommom :wflvmz o ocH mwv zuwmcoucm omcommoa suave: flom-oV m.H noofi-mv o.~m momcmu .wv xufimcuucH omcommmm cam: flmm-hu He Awn-mv we Amman“ ..muxv ow< cam: mofimsow NN ”magma Hm mmHmEom mo “monE um xom mm mHH mpuoflmmm UH; mpovcommouroz muoncoamom anhm~ .mHggHu nz< szzmzv m xHszmm< mmmozommmmzoz oz< mmmozommmm <2mmomhz< oz< Ohm mo zowum L ’l/ . _ ~Jawi. A?" ,m VM. ._ Frankie )‘4 14 J2? 4311.0. rown s V. Higgins, Graduate Student Professor "SC I . Mam Arm/Equal Win-fly hummu- 174 MICHIGAN STATE UNIVERSITY DEPAITHEVI' or mower ‘ NATURAL SCIEVCE 81.11056 EAST LANSING ' MICHIOAS - 0082‘ Winter, 1980 Dear Tamarisk Resident: Have you ever wondered why you like certain foods while other people you know find these same foods distasteful? For example, have you ever thought about why one person may like ketchup and relish on a hotdog while another person will eat only hotdogs with mustard and onions? If you were to take an opinion poll of different groups of people to deter- mine how many liked foods such as strawberries, asparagus, liver, tomatoes, spinach or other common fbods, you would discover a great variety of responses with respect to food preferences among the individuals of the different groups. As you probably know, such diversity of taste responses may be related to a nunber of factors such as the type of foods we eat most frequently or those foods which we have been influenced to like or dislike during our earlier years. What you may not know is that our perceptions of certain foods may be determined or strongly influenced by genetic factors--those same kind of factors which we have inherited from our parents that determine our blood type, eye color, height or other characteristics. Because the role of genetic factors in determining our differences in taste responses is not well understood, we are currently conducting a study of the genetics of taste perceptions and would be very grateful if you and your family would consent to be a part of this study. Your participation would involve the tasting of a few drops of several solutions, recording your taste perceptions and recording a few items of demographic importance such as age, sex, time of last food eaten, etc. This procedure, which can be done in your home, will only require about ten minutes per family member and has no greater risk than the tasting of common food substances. During the next few weeks a member of our team will be contacting you to schedule a convenient time for your family should you decide to participate in this study and answer any questions you may have regarding the project. If you prefer, you may complete and return the enclosed form as soon as possible to indicate your interest. (Please note: Because this is a genetic study, we are in need of families in which both mother and father are present in the household along with at least one child of age 7 or older, not including adopted children or children by previous marriages). We sincerely hope that you will consent to participate in this study which will help us learn more about those factors which determine individual taste preferences and responsiveness to certain food. Respectfull yours, Jgges V. Higginza Ph.D. Graduate Student Professor Telephone: 355-4600 Telephone: 353-2030 MSL'u am .4”le Art-ow "Equal Opportunity Immuno- 175 GENETICS OF TASTE PERCEPTION STUDY Please check appropriate responses below: We will participate in the Taste Perception Study. We may participate in the Taste Perception Study but have additional questions. We do not wish to participate in the Taste Perception Study for the following reason(s): Name Telephone Address Total number of non-adopted children in family Number of non-adopted children of age seven or older Best time to schedule our family: Weekday evenings Weekends (Please note: While we do hope that you will consent to participate in this study, it is important for accounting purposes that we hear from you even if you do not wish to volunteer. We would be most grateful if you would complete and return this fbrm in the envelope provided at your earliest convenience. Thanks in advance for your cooperation.) APPENDIX D CONSENT FORM AND SURVEY QUESTIONNAIRE APPENDIX D INFORMATION AND CONSENT FORM FOR PARTICIPANTS IN THE STUDY ENTITLED GENETIC STUDIES OF TASTE PERCEPTION OF ANTIDESMA AND PHENYLTHIOCARBAMIDE The study in which you are asked to participate may have future usefulness but at present is not essential to the diagnosis nor treatment of any known medical condition. it is a research study in which the differences in taste perceptions of Antidesma and Phenylthiocarbamide will be investigated. Antidesma is a fruit from which pies, jams, and Jellies are made and Phenylthiocarbamide is a substance often used in genetic studies of tasting. The purpose of this study is to gather information which may be useful for the improvement of our understanding of the inheritance of taste perceptions of these substances in human populations. it is important that you understand that no direct benefits to you are guaranteed by your participation in this study and that your responses will be kept confidential and that if published will be stated In such a way that anonymity will be preserved. You should also understand that the only acts required of you in this study are the taste sampling of certain solutions (antidesma, phenylthiocarbamide, quinine, fruit Juices. and OThOP common solutions) which should be of no greater hazard to you than the tasting of common food substances and the completion of a questionnaire to record your responses and other information of demographic Inporfance. You should further understand that you are free to discontinue your participation in this study at any time should you elect to do so. Statement of Consent The study entitled Genetic Studies of Taste Perception of Antidesma and Phenylthiocarbamide has been explained to me, and I understand the purpose, requirements and risks of my participation and freely consent to participate. i understand that in the unlikely event of physical injury resulting from research procedures, Michigan State University, its agents, and employees will assume that responsibility as required by law. Emergency medical treatment for injuries or illness is available where the injury or illness is incurred in the course of an experiment. l have been advised that i should look toward my own health insurance program for payment of said medical expenses. Signature Date if a minor (under age l8), parent or guardian must sign and state relationship. 1765 177 Questionnaire to be completed by Participants in the Investigation Entitled Genetic Studies of Taste Perceptions of Antidesma and Phenylthiocarbamide 1. Name 2. Date and Time of Test 3. Address 6. Phone Number 5. Student Number (if student) 6. Age 7. Sex (Circle One) Male Female 8. Race/Ethnic Group (Circle One) Hhite/Caucasian,fiBlack/Afro-American,4Chicano/Mexican American, Spanish American/Hispanic, American Indian, Asian/Pacific Islander 9. Time of last food eaten 10. Smoker or Non-Smoker (Circle One) 11. Taste Responses: Circle the term below which best describes the taste of the solutions listed then rate the intensity of that taste on a scale of 1-5 where #1 is mildest and 05 is strongest. (Example: If the solution tastes slightly salty, you would circle salty and the #1 next to the word salty). If the solution has no taste, circle the word tasteless. In each case where you can detect a specific taste, please complete the sentence following each solution to describe what that solution tastes most like from your previous taste experiences. Please be sure to eat an unsalted cracker and rinse your mouth after tasting each gin—ties- Solution A is: Solution g is: Tasteless Tasteless Mild '-—-€> Strong Mild ""5’ Strong Salty l 2 3 4 5 Salty l 2 3 4 5 Bitter l 2 3 4 S Bitter 1 2 3 4 5 Sweet 1 2 3 4 5 Sweet 1 2 3 4 5 Sour l 2 3 4 5 Sour 1 2 3 4 5 Solution A tastes like Solution g tastes like Solution 9 is: Solution 2 is: Tasteless Tasteless Mild --€’ Strong Mild ""'€> Strong Salty 1 2 3 4 5 Salty l 2 3 4 5 Bitter l 2 3 a 5 Bitter l 2 3 4 5 Sweet 1 2 3 4 5 Sweet 1 2 3 4 5 Sour l 2 3 4 5 Sour 1 2 3 4 5 Solution £_tastes like Solution 2 tastes like OVER 178 Solution §_is: Solution E is: Tasteless Tasteless Mild "‘9 Strong Mild -% Strong Salty 1 2 3 4 S Salty l 2 3 4 S Bitter l 2 3 4 5 Bitter l 2 3 4 5 Sweet 1 2 3 4 5 Sweet 1 2 3 4 S Sour l 2 3 4 5 Sour l 2 3 4 5 Solution E tastes like Solution {_tastes like Solution 9 is: Solution §_is: Tasteless Tasteless Mild ""5’ Strong Mild --€> Strong Salty l 2 3 4 S Salty l 2 3 4 5 Bitter l 2 3 4 5 Bitter 1 2 3 4 5 Sweet 1 2 3 4 5 Sweet 1 2 3 4 5 Sour l 2 3 4 5 Sour l 2 3 4 5 Solution E tastes like Solution 5 tastes like Solution 1 is: Solution £_is: Tasteless Tasteless Mild ‘9 Strong mm ---9 Strong Salty l 2 3 4 5 Salty l 2 3 4 S Bitter l 2 3 4 5 Bitter l 2 3 4 5 Sweet 1 2 3 4 5 Sweet 1 2 3 4 5 Sour 1 2 3 4 5 Sour l 2 3 4 5 Solution 1 tastes like Solution g_tastes like Solution 5 is: Solution L is: Tasteless Tasteless mm —-> Strong Mild ‘—> Strong Salty l 2 3 4 5 Salty l 2 3 4 5 Bitter l 2 3 4 5 Bitter 1 2 3 4 5 Sweet 1 2 3 4 5 Sweet 1 2 3 4 5 Sour l 2 3 4 5 Sour 1 2 3 4 5 Solution 5 tastes like Solution £_tastes like APPENDIX E RATIONALE FOR USE OF DIFFERENT STATISTICS EMPLOYED APPENDIX E RATIONALE FOR USE OF DIFFERENT STATISTICS EMPLOYED ' Data included in this dissertation have been analyzed by use of several statistics. For simple frequency data, ranges, means, medians and modes have been calculated where applicable. For comparisons of i one variable with another (crosstabulations), four statistical tests were employed to determine if associations existed between the variables. The tests used were the Chi-square, Cramer's V, Lambda Asymmetric, Lambda Symmetric and Z transformation statistics. These statistics were selected because they are more suitable when variables in cross- tabulation tables are measured at the "nominal" level, that is, variable values represent a distinct category and the value itself serves merely as a label or name fer the category (e.g., sweet, bitter, White/ Caucasian, male, female, etc.). Unlike ordinal-level and interval- level measurements, with nominal-level variables, no assumptions of ordering or distances between the categories are made. For analysis of family data and tests of genetic hypotheses, the Z-transformation and Fisher's exact probability test statistics were used. A brief description of each type of analysis used follows. 179 180 Chi Square The Chi-square test of statistical significance, used to deter- mine whether a systematic relationship exists between two variables is usually most appropriate when at least one of the variables can be placed into dichotomized categories (e.g., taster versus nontaster, bitter versus nonbitter, etc.), although in some instances this analysis is used when more than two categories for each variable are present. In crosstabulation tables, Chi-square is calculated by computing the cell frequencies which would be expected if pg relationship is present between the variables given the existing row and column totals. The expected cell frequencies are then compared to the actual values found in the table according to the fellowing formula: 1 . 2 2 = EIEO ' fi) 1 i f e X where f: equals the observed frequency in each cell, and f: equals the expected frequency calculated as where C1 is the frequency in a respective column marginal, ri is the frequency in a respective row marginal and N stands for the total number of valid cases. The greater the discrepancies between the expected and actual frequencies, the larger chi-square becomes. If no relationship exists between two variables in the sample under study, then any deviations from the expected values which occur 181 in a table based on randomly selected sample data are due to chance. While some small deviations can be reasonably expected due to chance, large deviations, i.e., large values of chi-square, are unlikely. Since we do not know what the actual relationship is in the universe, we interpret small values of chi-square to indicate the absence of a rela- tionship, often referred to as statistical independence. Conversely, a large chi-square implies that a systematic relationship exists between the variables. In order to determine whether a systematic relationship does exist, it is necessary to ascertain the probability of obtaining a value of chi-square as large or larger than one calculated from the sample, when in fact the variables are actually independent. This depends, in part, upon the degrees of freedom. The degrees of freedom vary with the number of rows and columns in the table, and they are important because the probability of obtaining a specific chi-square value depends on the number of cells in the table. By itself, chi-square helps us only to decide whether our vari- ables are independent or related. It does not tell us how strongly they are related. Part of the reason is that the sample size and table size have such an influence upon chi-square. Several statistics which adjust for these factors are available. When chi-square is thus adjusted it becomes the basis for assessing strength of relationship. Phi* For a 2 x 2 table, the phi statistic is a suitable measure of association, i.e., a measure of strength of relationship. Phi (¢) makes *This statistic is not used directly but its explanation is included here because the Cramer's V which is used is a modified version of Phi. 182 a correction for the fact that the value of chi-square is directly pro- portional to the number of cases N by adjusting the X2 value. Its fbrmula is: . n45 Phi takes on the value of 0 when no relationship exists, and the value of-+l when the variables are perfectly related, i.e., all cases fall just on the main or the minor diagonal. Cramer's V Cramer's V is a slightly modified version of phi which is suit- able for larger tables. When phi is calculated fer a table which is not 2 x 2, it has no upper limit. Therefore, Cramer's V is used to adjust phi for either the number of rows or the number of columns in the table, depending on which of the two is smaller. Its fbrmula is: ‘ i V = (min (fl, c-1)) V also ranges from 0 to +1 when several nominal categories are involved. Thus, a large value of V merely signifies that a high degree of associ- ation exists, without revealing the manner in which the variables are associated. Lambda Lambda is a measure of association fbr crosstabulations based on nominal-level variables. 183 Asymmetric lambda measures the percentage of improvement in our ability to predict the value of the dependent variable once we know the value of the independent variable. This is based on the assumption that the best strategy for prediction is to select the category with most cases (modal category), since this will minimize the number of wrong guesses. All the remaining measures of association are based on this concept, which is called proportional reduction in error. The fbrmula for asymmetric lambda is: where 2 mix. fjk represents the sum of the maximum values of the cell frequencies in each column, and max. f.k represents the maximum value of the row totals. The maximum value of lambda is 1.0, which occurs when predic- tion can be made without error, i.e., when each independent variable category is associated with a single category on the dependent variable. A value of zero means no improvement in predicting. Asymmetric lambda is computed for each of the variables. The two results are likely to be different since the one-way 0marginal) distributions are not usually the same. A symmetric lambda is also computed, which is a kind of average of the two asymmetric values. It makes no assumptions about which variable is dependent and it mea- sures the overall improvement when prediction is done in both direc- tions. Its fbrmula is: 184 22 mix. fjk + Z mJax. fjk - max. f.k - max. f3. symm 2N - max. f.k - max. fj Lambda = k where 2 max. fjk and max. f.k are as defined fer lambda asymmetric, k max. f5 is the maximum column total, and 2 max. fjk is the sum of the maximum values of the cell frequencies in each row (Nie st 31., 1975). Z-transformations (Transformation to Standard Normal Distribution) For testing dominant-recessive hypotheses by use of Snyder's ratio, Z-transformations are more appropriate since Snyder's ratios calculate the expected proportions of offspring from the various matings. Chi-square analyses are less applicable here since the dif- ferences between two proportions are compared and because in some instances the expected proportion of certain types of offspring is zero. For example, fer the hypothesis that the inability to taste PTC is recessive, the expected proportion of offspring with the dominant taster phenotype from nontaster x nontaster matings is zero. If Chi- square analysis is used data from this mating combination cannot be tested. Furthermore, Chi-square analysis requires use of whole numbers rather than proportions. The Z transfbrmation however, can be used to test the difference between two proportions as well as allow the use of data from all mating types in acceptance or rejection of the hypotheses under consideration and is computed as fellows: Z = Obs. - Exp. /Ob$ ell’Obs e) N 185 where Obs. = observed proportion of offSpring of a given type from a particular mating Exp. = expected proportion of offspring of a given type from a particular mating N = total number of offspring from the particular mating. Probabilities of 2 values thus obtained are then determined from Cumulative Standard Normal Distribution Function Tables in the form of F(Z). If Z values are positive then 1 - F(Z) =<1/2. For negative Z values, F(Z) =<1/2. From the