Population variation in asymmetry and diversity from finger to finger for digital ridge-counts.код для вставкиСкачать
Population Variation in Asymmetry and Diversity from Finger to Finger for Digital Ridge-counts RICHARD L. JANTZ Department of Anthropology, University of Tennessee, Knoxville, Tennessee 37916 KEY WORDS tion Dermatoglyphics . Asymmetry. . Ridge-counts . Racial varia- ABSTRACT Population variation in ridge-count asymmetry and diversity from finger to finger has received scant attention in dermatoglyphic studies. Asymmetry, in particular, has generally been attributed to environmental effects operating during the formation of dermal ridges. Examination of samples from several groups of diverse racial background revealed the existence of considerable population variation with respect to finger ridge-count asymmetry and diversity from finger to finger. Patterning along population lines suggests a genetic rather than environmental basis for such variation. The genetic mechanisms responsible for ridge-counts may also mediate asymmetry and diversity, or the degree of developmental stability in different populations may itself be under genetic control. It has long been recognized that homologous digits of the same individual often display different dermatoglyphic patterns (Cummins and Midlo, '61). The consistency in levels of asymmetry among diverse groups led these authors to conclude that there was little in the way of racial variation with respect to pattern type asymmetry. More recently a number of investigators have examined bilateral asymmetry with respect to ridge-counts. Asymmetrical ridge-counts have been attributed to "developmental indeterminancy" (Rothammer et al., ' 7 3 ) , and increases i n asymmetry have been observed in chromosomal mosaics (Polani and Polani, '69) and cleft palate patients (Adams and Niswander, '67). Family studies have indicated that asymmetry lacks a genetic basis entirely (Holt, '54) or has only a small component (Singh, '70). Likewise, it has long been recognized that ridge-counts on non-homologous digits are correlated but may show considerable variation in some individuals. Racial variation has been shown to exist with respect to ridge-count diversity from finger to finger (Jantz, '74), and Holt ('60) demonstrated a substantial genetic comAM. J. PHYS. ANTHROP., 42: 215-224. ponent in her family material. Neither asymmetry nor diversity from finger to finger has ever been systematically considered from the standpoint of population variation despite the information that might result from such a n approach concerning the genetic and developmental nature of dermatoglyphics. The aim of this study is to investigate ridge-count asymmetry and diversity among non-homologous digits in a number of diverse groups of mankind. The extent of racial and population variation is explored along with relationships among asymmetry, diversity from finger to finger and ridge-counts. MATERIALS AND METHODS A number of different measures have been used to quantitatively evaluate asymmetry in finger ridge-counts. Holt ('54) used summed ridge-counts of right and left hands, Parsons ('64) the summed absolute differences between homologous digits and Singh ('68) the mean of the squared differences between homologous digits. I n any measure of asymmetry it would seem desirable to square differences between homologous digits to emphasize the larger and presumably more important 215 216 RICHARD L. JANTZ asymmetries. For present purposes, the following measure of asymmetry seems appropriate : i = l Taking the square root of squared differences will result in a more symmetrical distribution than simply using A'. Holt ('58) evaluated diversity from finger to finger by computing the quantity : i=l where q, is the ridge-count for the ith digit and Q is Lq, or the total ridge count. Later Holt ('60) used the measure S/ \/lOor the standard deviation of counts over 10 fingers to study familial correlations. It is apparent that ridge-count asymmetry a s well as diversity among non-homologous digits will contribute to S/\/lo. The effect of asymmetry may be easily eliminated by summing the ridgecounts of homologous digits so that only diversity among non-homologous digits contributes to the value. Holt's measure then becomes : s / d 5 = j F Computation of these two values for each individual provides convenient, independent, numerical expressions of that individual's degree of asymmetry and diversity among non-homologous digits. Population characteristics can then be determined and comparisons made in the usual manner. The samples available for study consist of two groups of European extraction (English and American White), four of African extraction (Dogon, Efe Pygmy, Bedik-Bassari and American Negro) and one Polynesian group (Easter Island). Ideally, one would want to have samples representing the major groups of mankind, but the present samples comprehend virtually the total known range of ridgecount variation in man from the lowest (Pygmy) to the highest (Easter Island). The American White sample (133 males and 132 females) was drawn from the student body of the University of Tennessee. Only students who gave their birthplace as Tennessee or adjacent states were included in the sample so that it may be considered broadly representative of the White population of the Southeastern United States. No information as to the ethnic composition was obtained, but the majority of the subjects are probably of British Isles extraction, since the Southeast was settled mainly by people of Scottish, Irish and English descent. Ridgecounts and S f i data for this sample appear in Jantz ('74). The English sample (151 males and 151 females) consists of the parents of Dr. S. B. Holt's large family sample. Only parents were included to eliminate related individuals and to keep the sample size on more or less the same order as the other groups in the study. This sample is broadly representative of the population of Southeastern England and in particular, the London area. Ridge-count data have been reported in Holt ('58, '68) and S/\/10data appear in Holt ('58) and Jantz ('74). American Negro (102 males, 122 females) dermatoglyphic prints were obtained principally from students in elementary and junior high schools in Knoxville, Tennessee, but a few came from the University of Tennessee as well. Most of the subjects were born in Knoxville, but many had parents who came to Knoxville from many other areas in the Southern United States. This sample can be taken as representative of Southern Blacks, generally, and not of the Black population of Knoxville in particular. Ridge-counts and S / f i data have been presented in Jantz ('74). Efe Pygmy (152 males and 53 females) and Dogon (169 males and 100 females) prints are those located in the Institute of Human Biology, Utrecht, Netherlands. The Pygmy prints were collected in the 1930's by Dr. P. Julien, and ridge-counts of this sample are presented by Glanville ('69). Dogon dermatoglyphics were collected during the Tellem Expeditions of 1964-1965, conducted by the Institute of Human Biology, Utrecht. Finger dermatoglyphics of this sample have been presented by Huizinga ('65) and Glanville and Huizinga ('66). 217 RIDGE-COUNT ASYMMETRY AND DIVERSITY Bedik-Bassari (103 males and 55 females) are two related tribal groups located in Eastern Senegal. Dermatoglyphic prints for both groups are located in the MusCe de l'Homme, Paris. The dermatoglyphics of the Bedik have been presented by Gomila et al. ('65); those of the Bassari are as yet unpublished. Data for these two groups have been pooled for purposes of the p r e s e n t analysis. Ridge-count and S / d l O data for this sample appear in Jantz ('74). Easter Island (146 males and 141 females) dermatoglyphic prints were collected by Dr. R. J. Meier during the course of his human biological investigations. Meier ('69) identified approximately 14% of his sample as containing no recent admixture, 39% were mixed with other Polynesians, and 47% possessed some European admixture. An analysis of the Easter Island dermatoglyphics appears in Meier ('69). All ridges were counted according to methods set forth in Holt ('68), the larger count being used in the case of whorls. Ridge-counts of American Whites and Negroes and Bedik-Bassari were made by the author; for the remainder the counts made by previous investigators were made available to me. Intergroup differences were evaluated by calculating the standard error of the difference between means for each pairwise comparison as described by Fisher ('58). Results may then be presented as normal deviates, with probabilities obtained from a normal curve table. This procedure is based on the assumption that variables being tested are normally distributed, a requirement not met by the measures of diversity and asymmetry used in this study. However, the distribution of differences between means will frequently approximate the normal distribution more closely than the variables themselves, so results obtained in this manner should be acceptable for present purposes. RESULTS The means and standard deviations for total ridge-count (TRC), S / 6 a n d m are presented in table 1. TRC values show no remarkable features and are in accord with previous concepts of population variation. It is apparent that neither S/V% nor *means bear a systematic relationship to the pattern of variation established by TRC. In general the Ne ro populations have the lowest S 6 a n d A' values, indicating greater homogeneity among fingers, and less asymmetry, as well as the lowest ridge-counts. However, the Easter Islanders, with the highest ridge-counts, are characterized by s/= and values only slightly above those of the Negroes, while English and American Whites show the greatest inter-finger heterogeneity and asymmetry. The magnitude of population differences for all three measures is indicated by the normal deviates for each painvise comparison in table 2. The two groups on the extremes of the ridge-count range, Easter Islanders and Pygmies, differ significantly from all other groups, and of course from each other, with respect to TRC. English males differ from the various Negro groups, but beyond that, significant differences are few. For samples of the size used here, it is difficult to demonstrate significant differences for groups falling in the intermediate ridge-count range. The same is F * TABLE 1 Means and standard deviations f o r TRC, S/\/rand 4Z Males Means TRC Sldd Females Standard deviations dm Means TRC SId3 dxi TRC 46.93 55.74 45.71 47.25 45.32 50.11 49.40 2.79 3.29 3.20 2.79 2.78 2.83 2.59 4.07 4.07 3.98 3.14 3.91 3.37 3.62 162.30 127.87 126.08 129.21 118.44 122.95 97.66 S / d s fl Standard deviations TRC SldS di@ 52.22 51.37 50.02 47.05 47.46 28.50 49.92 3.09 3.35 3.22 2.97 3.68 3.21 2.73 3.84 4.27 4.22 3.54 3.33 3.28 4.00 ~~ Easter Island English Am. White Am.Negro Dogon Bedik-Bassari Pygmy 175.53 145.76 134.74 130.81 126.46 121.55 96.65 6.83 8.17 7.88 9.20 7.35 8.98 5.96 7.14 6.51 8.06 5.72 7.02 5.55 7.41 6.54 7.18 6.94 6.27 6.82 6.22 5.49 7.93 9.10 8.00 7.97 8.10 7.42 7.69 218 RICHARD L. JAN’IZ a, true for S / 6 and with only the groups on the extremes of the ranges, English and American Whites on the one hand, and the Negro samples on the other showing consistently significant differences. It is of considerable interest to note the close correspondence between the means of S / 6 and among the different samples. As can be seen from figure 1, groups with high S / d 5 values are also the most asymmetrical, and conversely, groups with greatest symmetry show the greatest amount of homogeneity among non-homologous fingers. This is the more surprising since it is not apparent in individuals on the intrasample level. Correlation coefficients between and S/ G r a n g e from 0.02 in Dogon females, to 0.40 in American White females, with most of the values falling between 0.20 and 0.35. This might suggest that common factors are responsible for asymmetry and inter-finger variation, but that they tend to take one form or the other in the individual, culminating in similar values at the population level. Sex differences for TRC, S e and are shown in table 3. None of the differ- + + t Ap I 0 Females t m + + t I 6.0 1 I 7.0 8.0 S/E Fig. 1 Bivariate plot showing the relationship between mean values of S/&and fi for each sample (abbreviations: E.I., Easter Island; E, English; AW, American White; AN, American Negro; D, Dogon; BB, Bedik-Bassari; P, Pygmy). ences is statistically significant except those for TRC in Easter Island and English samples. In most human populations males have the larger ridge-counts, and that bias exists in the present samples, except for a TABLE 2 Normal deviates of all pairwise comparisons f o r TRC, S/* TRC Comparison E. Island X English E. Island x Am. White E. Island x Am. Negro E. Island X Dogon E. Island x Bedik & Bassari E. Island x Pygmy English X Am. White English x Am. Negro English x Dogon English x Bedik & Bassari English x Pygmy Am. White x Am. Negro Am. White X Dogon Am. White x Bedik & Bassari Am. White x Pygmy Am. Negro x Dogon Am. Negro x Bedik & Bassari Am. Negro x Pygmy Dogon x Bedik & Bassari Dogon x Pygmy Bedik & Bassari X Pvnmv 1 2 P < 0.05. P < 0.01. 3P < 0.001. Males 4.9g3 7.35 7.35 9.403 8.59 14.123 1.83 2.29 3.37 3.61 8.11 0.64 1.57 2.08 6.76 0.74 1.36 5.55 0.81 5.61 3.91 S N 5 Females Males Females 5.67 5.85 5.40 6.78 6.73 7.93 0.30 0.22 1.49 0.87 3.763 0.51 1.19 0.58 3.503 1.68 1.09 3.91 0.74 2.49 3.21 2.96 1.45 2.42 1.01 3.08 4.11 1.37 4.98 3 3.99 5.60 6.85 3.55 3 2.40 4.16 5.18 1.58 0.62 1.19 2.27 3.22 0.41 1.70 1.03 0.73 0.60 0.65 2.32 0.63 2.38 0.81 . 1.89 3.67 1.72 0.26 1.40 3.10 2 1.19 0.11 1.70 1.05 2.53 2 1.27 ’ ~~ and fiz Males 2.19 1.68 2.25 0.26 2.44‘ 1.69 0.46 4.55 3 2.57 4.67 1.69 3.973 2.02’ 4.103 3.463 2.12’ 0.26 0.64 1.13 1.53 0.89 dD Females 2.47 ’ 1.76 0.07 0.53 0.93 0.37 0.61 2.41 1.92 2.99 0.37 1.70 1.24 2.39 1.65 0.47 0.99 0.42 1.37 0.75 0.38 219 RIDGECOUNT ASYMMETRY AND DIVERSITY the Pygmies and Bedik-Bassari. S/ and sex differences do not parallel those of TRC, and i t is noteworthy that in all the Black samples, except for S / a in Pygmies, the females exceed the males, while the reverse is true among Easter Islanders, English, and American Whites. These last three are characterized by the greatest sex differences in TRC. While no sex difference in ~ / G o d r F is significant, the consistent patterning suggests that this reflects a biological reality rather than sampling fluctuations, and kurtosis (gz) for the S/ s a n d A2 distributions are presented in table 4. Both variables are characterized by a strong positive skew and a mild leptokurtosis. Most gl values are significantly different from zero, and all are positive. Fewer of the g2 values attain statistical significance, but those which do are all positive. The shapes, of these distributions contrast with those usually encountered for TRC, which are negatively skewed and flattened (Holt, '55; Jantz, '74). Finally, it is of interest to examine the relationship between TRC, S/ 6and The correlation coefficients of TRC with S/ G a n d are shown in table 5. About half attain statistical significance, and all but those for Pygmies, Easter Island males, and American Negro males for TRC and are negative. The generally negative nature of these correlations agrees with that obtained by Holt ('58) between TRC and S2,and between TRC and S/mfor American Whites, American Negroes, and Bedik and Bassari (Jantz, '74). The contribution of each of the digits to S/ 6and was ascertained by computing the correlation coefficient of each digit with the asymmetry and diversity values. These results are set forth in figure 2. Most striking is the strong contribution made by certain fingers, and the apparent absence of relationship on the part of others. There is in fact, remarkable conformity both by sex and group concerning the attern of correlations, particularly for S/ 5. For this variable, most groups display negative correlations on the order of 0.35-0.45 for digits I1 and 111, and correlations of approximately zero for the remainder of the digits. The Pygmies and Easter Islanders do not show negative correlations i n digits I1 and 111, but instead display elevations in the other digits, thus preserving the characteristic shape of the curve, with dips corresponding to digits I1 and 111. It would appear then, that the level of inter-finger diversity in a n individual results from interaction between two groups of digits, with digits I1 and I11 behaving as one unit, and I , IV, and V as m, SkewnessP) T TABLE 3 Sex differences ( m a l e - f e m a l e ) f o r T R C , S / v 3 and diF Easter Island English Am. White Am. Negro Dogon Bedik-Bassari PY gmy 1 P ZP + + + + 0.29 +0.70 +0.41 -0.31 - 0.30 - 0.50 +0.06 13.23 1 +17.89 8.66 1.60 8.02 - 1.40 - 1.01 + +0.24 +0.10 4-0.18 - 0.83 -0.12 - 0.40 -0.28 < 0.05. < 0.01. TABLE 4 Skewness (gl) and kurtosis (92) for S / & and dAZ Skewness (gl) s/fi Group E. Island English Am. White A m . Negro Dogon Bedik-Bassari Pygmy 1 2 3 P P P < 0.05. < 0.01. < 0.001. Kurtosis (g2) vm v= S / G Males Females Males Females Males Females 0.731 0.338 0.723 0.838 0.533 0.951 0.429 0.868 0.726 0.423 1.268 3 1.226 1.106 0.409 1.1003 0.467 0.720 0.729 1.000 1.243 3 0.746 1.648 0.947 0.876 0.634 2 0.627 0.435 0.774 0.413 -0.359 0.678 0.327 0.245 0.473 0.214 1.1462 0.453 -0.415 2.621 3 1.807 0.926 0.225 Males Females 1.619 3 6.555 3 -0.479 1.587 3 0.215 0.395 0.690 -0.158 1.278 0.226 1.662 - 0.870 0.982 1.718 220 RICHARD L. JANTZ TABLE 5 Correlation coeficients of TRC w i t h S/dg and d z Group Easter Island English Am. White Am. Negro Dogon Bedik-Bassari Pygmy 1P 2 TRC and S / d 5 Males TRC and VAT Females Males -0.02 -0.21' -0.03 -0.25 -0.20 -0.34 -0.10 -0.2S3 -0.05 - 0.22 - 0.13 +0.18' +0.25 +0.14 -0.12 -0.17 +0.01 +0.05 Females -0.19 -0.06 -0.14 -0.18 - 0 . 1 6 ' -0.06 - 0.31 3 - 0.331 4-0.19 +0.02 < 0.05. < 0.001. P < 0.01. 3 P the other. Disparity in ridge-counts between these groups tends to produce high values of S/ 6 Correlation of the individual digits with is less clearly patterned than is the case in S / a . Here too, however, digit 111, and to a lesser degree, digit I1 are negatively correlated with asymmetry, so that the pattern is similar to that of S\/Js although marked by less fluctuation. DISCUSSION The data presented in this paper suggest that diversity from finger to finger and asymmetry between homologous fingers, have considerably greater biological significance than has previously been accorded them. In these samples the magnitude of inte opulation differences revealed by S / 5 and approaches that for TRC, but the patterning of interpopulation differences is quite different. These results also raise some interesting questions concerning the underlying causes of asymmetry and inter-finger diversity. First, it is useful to consider the present results in the light of the limited number of studies dealing with the same topic. So far as I am aware, a study by Salzano and Benevides ('74) constitutes the only other instance where racial differences in ridge-count asymmetry were sought. Their samples, however, consisting of Brazilian Whites, and Brazilian Blacks with various degrees of white admixture, failed to show any significant differences in asymmetry, and no systematic influence of white admixture. Their samples did show generally lower asymmetry in Negroes, and i t is per- \;p haps important that mean TRC values were very similar in all of their groups. Studies concerned with diversity among non-homologous fingers are likewise scarce. In a previous study using the measure of diversity S/ I demonstrated the existence of interpopulation variation among American Whites, American Blacks, English, and Bedik-Bassari. Sunderland and Coope ('69) and Basu ('72) calculated Holt's S' for Jivaro and Mysore Brahman samples respectively, but the results are not presented in such a way a s to facilitate comparison with present data. With such limited comparable information, it is sufficient simply to point to the existence of considerable intergroup heterogeneity, and a more general picture of the extent of variation within and among the major groups of man must await further investigations. For purposes of assessing the racial significance of inter-finger diversity and asymmetry it is desirable to have information concerning their genetic and environmental components. Such information can be obtained from family studies, but here too we are faced with a paucity of information. Holt's ('60) familial correlations for S / m reveal a substantial genetic component, and interpopulation variation apparent among the present samples is explicable in those terms. Asymmetry, however, while showing some hereditary influence, apparently has a high environmental component (Singh, '70). This is the case as well for dental asymmetry (Bailit et al., '70) but unlike dental asymmetry it is not possible to suggest health, nutritional, or other environmental circumstances as responsible for dermatoglyphic asymmetry. For example, American Negroes, whose environment is no doubt very different from that of African Negroes, are most similar to Africans with respect to both and S / G . Levels of inter-finger diversity and asymmetry appear to be patterned along population lines rather than related to any identifiable environmental factor. Is it possible to relate the observed interpopulation variation to existing hypotheses concerning possible underlying genetic mechanisms? Two possible explanations suggest themselves, neither of which can m, 22 1 RIDGE-COUNT ASYMMETRY AND DIVERSITY +4 -4 +4 - - Male S/6 r ---- Female r English -2 -4 -6 ( 1 1 1 1 1 1 1 1 1 1 L I I I I I I I I I +2 r American White -2 -4 c Q) .- -6 , I / . , c .2 -2 r'\ \--- -4 ~ 0 , \ -2 -4 +4 , '\ ' , \I/ \---' II 'I ' ,-. . , \ Bedik +6 r Digit Fig. 2 Correlation coefficients of individual digits with S/& ordinate is X 10. Digit and m. Scale of the 222 RICHARD L. JANTZ be accepted or rejected on present evidence. Holt (’58) suggested, in her study of the relationship of S’ to TRC, that the TRC values showing the highest Sz might be expected to consist of individuals most heterozygous in the polygenic system responsible for TRC. If one assumes that the genetic mechanism responsible for ridgecounts in different populations is the same, apart from differences in frequency of the various alleles, then the proportion of heterozygotes, or more accurately, the ratio of heterozygotes to homozygotes, will vary accordingly. Although we know little about gene action concerning ridge-counts, it is at least reasonable to suggest that individuals heterozygous at the most loci will display the highest levels of asymmetry or inter-finger diversity. Looking at the data from this standpoint, the Pygmies, who presumably have the highest frequency of genes resulting in low ridge-counts, rank lowest with respect to S / f i and As ridge-counts increase there is generally a concommitant rise in these values, until they reach a maximum in the English, followed by a decline again in Easter Islanders, where the highest ridge-counts are found. Easter Islanders presumably have high frequencies of those alleles resulting in high ridge-counts, and thus fewer heterozygotes. Groups such a s English or American Whites, under this assumption, would have intermediate allelic frequencies, and as TRC increases or decreases, so also does the fraction of homozygotes, reflected in greater homogeneity among the ten fingers. The low heritability just noted for asymmetry might also be explained in these terms, since genotypes homozygous for contrasting pairs of alleles would be grouped into the same phenotypic class for purposes of computing familial correlations. Another possibility relates to genetic control of developmental stability, or canalization. Waddington (’60) has shown that some Drosophila genotypes possess greater abilities to withstand environmental insults than others, so that canalization itself is under genetic control. Mather’s (’53) experiments with asymmetry in sternoplueral chaetae number in Drosophila revealed evidence for a genetic component controlling developmental asymmetry, and dif- ferent Drosophila lines showed different degrees of asymmetry which were independent of environment. It is quite likely that similar mechanism exist which would result in different degrees of developmental asymmetry in man. As far as the present samples are concerned, it is perhaps significant that groups showing greatest homogeneity among digits are, or recently were, tribal populations, where environmental stresses might be expected to be greater than those experienced by groups with a long history of civilization. Under such circumstances genotypes better able to resist environmental insults would be at a selective advantage, resulting in greater developmental stability for such groups. It is not possible on present evidence, to say which, if either, of these two hypotheses is the most tenable, and for that matter they are not necessarily mutually exclusive. Appropriate family material, and additional population data are necessary before attempting to carry these ideas any further. It is clear from present evidence however, that relationships among digits are just as important as the number of ridges as far as population variation is concerned. ACKNOWLEDGMENTS I am grateful to the following people for making dermatoglyphic prints or data available to me: Dr. S. B. Holt (English), Professor R. Gessain and Madame M. Gessain (Bedik-Bassari), Professor J. Huizinga (Dogon and Pygmy) and Dr. R. J. Meier (Easter Island). I am also grateful to Dr. Roy Wallace, Superintendent of Instruction, Knoxville Public Schools; to the principals and students of the several elementary and junior high schools, and to the students at the University of Tennessee for their cooperation in this study. My appreciation for much helpful discussion goes to Dr. Sarah Holt and Professor C. A. B. Smith of the Galton Laboratory. This investigation was supported by a National Institutes of Health Fellowship No. 1 FO GM55508-01 from the National Institute of General Medical Sciences. LITERATURE CITED Adams, M. S., and J. D. Niswander 1967 Developmental “noise” and a congenital malformation. Genet. Res., 10: 313-317. RIDGE-COUNT ASYMMETRY AND DIVERSITY Bailit, H. L., P. L. Workman, J. D. Niswander and C. J. MacLean 1970 Dental asymmetry as an indicator of genetic and environmental conditions in human populations. Hum. Biol., 42: 626-638. Basu, A. 1972 Quantitative dermatoglyphics in Mysore Brahman. (Abstract). Int. Dermatoglyphics News Bull., 1: 4. Cummins, H., and C. Midlo 1961 Finger Prints, Palms and Soles. Dover, New York. Fisher, R. A. 1958 Statistical Methods for Research Workers. Oliver and Boyd, London. Thirteenth edition. Glanville, E. V. 1969 Digital ridge-counts of Efe Pygmies, Am. J. Phys. Anthrop., 31: 427- 428. Glanville, E. V., and J. Huizinga 1966 Digital dermatoglyphics of the Dogon, Peul, and Kurumba of Mali and Upper Volta. Proc., Koninkl. Nederl. Akad. Wetensch., Series C, 69: 664-674. Gomila, J., L. Pee-Laborde and M. Th. de Lestrange 1965 Dermatoglyphes digito-Palaires et plis de flexion d a m l’isolat Bedik, Senegal Oriental (resultats prelimiaires). Proc. VIII Czeckoslovak Anthrop. Congr. Bmo, 19 (11): 103-118. Holt, S. B. 1954 Genetics of dermal ridges: bilateral asymmetry in finger ridge-counts. Ann. Eugen., 18: 211-231. 1955 Genetics of dermal ridges: frequency distribution of total finger ridge-count. Ann. Hum. Genet., 20: 159-170. 1958 Genetics of dermal ridges: The relation between total ridge-count and the variability of counts from finger to finger. Ann. Hum. Genet., 22: 23Z341. 1960 Genetics of dermal ridges: familial correlations for S/glO, a measurement of the 223 diversity of ridge-counts from finger to finger. Ann. Hum. Genet., 24: 253-269. 1968 The Genetics of Dermal Ridges. Charles C Thomas, Springfield. Huizinga, J. 1965 Finger patterns and ridge counts of the Dogons. Proc., Koninkl. Nederl. Akad. Wetensch., Series C, 68: 398-411. Jantz, R. L. 1974 Finger ridge-counts and interfinger variability in Negroes and Whites. Hum. Biol., 46: 663-675. Mather, K. 1953 Genetical control of stability in development. Heredity, 7: 297-336. Meier, R. J. 1969 The Easter Islander: A Study in Human Biology. Ph.D. dissertation, University of Wisconsin, Madison. Parsons, P. A. 1964 Finger print pattern variability. Acta Genet., 14: 201-211. Polani, P. E., and N. Polani 1969 Chromosome anomalies, mosaicism and dermatoglyphic asymmetry. Ann. Hum. Genet., 32: 391-402. Rothhammer, F., J. V. Neel, F. da Rocha, and G. Y. Sundling 1973 The genetic structure of a tribal population, the Yanomama Indians. VIII Dermatoglyphic differences among villages. Amer. J. Hum. Genet., 25: 152-166. Salzano, F. M., and F. R. de Sa F. Benevides 1974 Fingerprint quantitative variation and asymmetry in Brazilian Whites and Blacks. Am. J. Phys. Anthrop., 40: 325-328. Singh, S. 1968 A measure of asymmetry of finger ridge-counts. Acta Genet., 18: 599-605. 1970 Inheritance of asymmetry in finger ridge counts. Hum. Heredity, 20: 403- 408. Sunderland, E., and E.Coope 1969 The Jivaro: quantitative digital dermatoglyphics. Man, 4: 438-442. Waddington, C.H. 1960 Experiments on canalizing selection. Genet. Res., 1: 140-150.