AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 57:99-lOl(1982) Dermatoglyphic Characters and Physique: A Correlation Study FRANCISCO ROTHHAMMER, ELENA LLOE: AND JAMES V. NEEL Departamento de Biologia Celular y Genetica, Universidad de Chile, Santiago, Chile (F.R.,E.L.),and Department of Human Genetics, University of Michigan Medical School, Ann Arbor, Michigan 48109 (J.KN.) KEY WORDS Dermatoglyphics, Hand measurement, Early developmental interaction ABSTRACT The association of anthropometrics, particularly hand measurements, with dermatoglyphic characters is quantified. Children with square hands exhibit higher main line indices, a-b ridge counts, and more open atd angles. Adults with broader hands have more arches. Taller individuals with larger hands present higher a-b ridge counts and leaner subjects with long narrow hands, closer atd angles. The correlation of physique and dermatoglyphics is small but if verified, suggests that at early fetal stages, factors responsible for the establishment of dermatoglyphic patterns interact with genetic determinants of adult shape that are already active. The question of whether the observed phenotypic variability in dermatoglyphic traits reflects an underlying genetic variability has been the object of many presentations [see, for example, Wichmann (1963)and Holt (1968)for references]. Previous limited studies by Tillner (1942), Wendt (1954), and Wichmann (1963) have revealed small and inconsistent correlations between various aspects of size and shape and total ridge count andlor pattern intensity. Stimulated largely by our observation of an apparently loose association between various indicators of mean bodily size and the index of pattern intensity (and in ignorance of the studies just quoted), we undertook the present study, which represents an effort to quantify the association, if any, between dermatoglyphic characters and some pertinent body measurements. SUBJECTS AND METHODS A sample of 105 unrelated adults, 25 males and 80 females and 108 children, 60 males and 48 females, descendants of the former, were obtained in the northern area of the socioeconomically middle-classpopulation of Santiago.Correa et al. (1976)sampled the same population for their blood group study. The proportion of Spanish genes in this population has been estimated to be 59% (Valenzuelaet al., 1978). The ages of the adults lie between 26 and 46 years 0002-948318215701-0099$01.50 0 1982 ALAN R. LISS, INC. and those of the children between 5 and 9 years. The following observations were obtained for each individual: breadth and length of the middle finger, breadth and length of the hand, stature, weight, and finger and palmar dermatoglyphic prints using the Faurot technique. Measurements were obtained in the manner described in the International Biological Programme Handbook (Weinerand Laurie, 1969: 8-16). The middle finger was selected for detailed measurements because of its relevance for measuring hand length. Palm length was obtained as the difference between hand length and finger length. The following shape measurements were included in the analysis: hand index (hand breadthihand length), palm index (hand breadthipalm length), and ponderal index (statureivweight). Among the dermatoglyphic traits considered were pattern intensity index, a-b ridge count, atd angle, main line index, and c line absence. Age dependent anthropometric and dermatoglyphic variables (atd angle among the latter) of the children were adjusted to the mean age by quadratic polynomial regression. Adult measurements were not adjusted for age. Systematic sex differences in anthropometric and dermatoglyphic variables were removed by linear regression. Right and left hand measureReceived April 3, 1981: accepted August 25. 1981 F. ROTHHAMMER, E. LLOP, AND J.V. NEEL 100 ments were averaged. Finally, the associations of the adjusted anthropometric and dermatoglyphic variables were quantified by computing Pearson’s product-moment correlation coefficients. A function of r, namely ment. Pattern intensity index is, for example, associated negatively with hand width, main line index is correlated negatively with finger length, a-b ridge count is positively correlated with hand size measurements and stature, and, finally, atd angle is associated with hand measurements and ponderal index. where n is the sample size, and which is distributed as t (Mood and Graybill, 1963),was evaluated and referred to the cumulative “Student’s” distribution to test the null hypothesis Q = 0. In mixed population samples, as the present one, two traits may appear correlated as a r e sult of “racial stratification.” None of the dermatoglyphic traits studied exhibit a clear-cut gradient of distribution among populations with different percentages of Amerindian admixture. Stratification seems therefore unlikely as an explanation of the “significant” correlations observed. However, in 9 x 5 matrices such as Table 1 and 2, with a total of 90 correlations, one should expect approximately 5 of the correlations to be “significant” on the basis of chance alone. In fact, there are 15 “significant” correlations, but the largest is only 0.31. Since Tables 1 and 2 represent the same measurements at different stages in the life cycle, it would be reasonable and conservative to expect r e d y significant correlations to be duplicated in the two tables, although there DISCUSSION RESULTS The presentation of descriptive statistics such as means or variances is irrelevant for the purposes of this study and therefore omitted. Correlations are summarized by sex in Tables 1and 2. Inspection of the children’scorrelation matrix (Table 1)reveals that main line index, a-b ridge count, and atd angle are significantly associated with the hand index and the latter two measurements, also with the palmar index. The adult correlation matrix (Table 2) seems to exhibit a somewhat different structure, probably as a result of the size and shape changes that occur during growth and develop- T A B L E 1. Children correlation matrix for various anthropometnc and dermatoglyphic characteristics Characteristic Finger width Finger length Hand width Hand length Stature Weight Palmar index Hand index Ponderal index Pattern intensity index Main line index ~ .06 .oo .07 .12 -.18* .09 .13 .12 .04 .04 .oo .03 -.15 -.13 -.09 .ll .19* -.07 a-b ridge count .16 .13 .16 .01 .08 .09 .26** .18* -.02 c-line absence atd angle .10 .06 -.02 -.04 .03 .17* -.01 -.08 -.01 .07 .13 .13 .10 .12 .24* .19* -.04 -.11 * p < 0.05. ** p < 0.01. T A B L E 2. Adult correlation matrix for various anthrouometric and dermatoplvuhic characteristics ~ Characteris tic Finger width Finger length Hand width Hand length Stature Weight Palmar index Hand index Ponderal index * p < 0.05. ** p < 0.01. Pattern intensity index -.06 .06 --.19* .06 -.04 .06 - -.15 -.13 -.08 Main line index - .06 -.18* .03 - .08 .04 .ll -.01 .13 - .08 a-b ridge count .05 .31** .28** .25** .21* .15 .15 .07 -.02 ~~ atd angle ~~ c-line absence -.07 -.15 -.02 -.20* - .04 -.14 -.16 -.07 .11 .13 .14 .17* -.22* -.07 -.16 .06 .10 -.15 DERMATOGLYPHIC CHARACTERS AND PHYSIQUE is of course no absolute a priori reason that this be the case. This is true for only two of the entries, involving negative correlations between main line index and finger length and positive correlations between hand index and atd angle. We take these to be valid findings. In addition, there is a cluster of positive significant correlations in the adult between a-b ridge count and measures of finger and hand size which, while not significant in the children, tend d s o to be positive. Finally, we believe that the correlation between palmar index and a-b ridge count is probably meaningful. In descriptive terms, big square hands are characterized by transversally aligned palm ridges, more open atd angles, and higher a-b ridge counts. Notable, if verified, is the correlation between hand width and pattern intensity in adults, since it suggests that the number of triradii on the fingertips may be related to hand morphology. The only point of cross-contact with previous work with which we feel comfortable is Wichmann’s (1963) report of very small (very nonsignificant)positive correlations of stature with ridge count, with which our findings do not disagree. The existence of these small correlations raises interesting considerations with respect to genetic factors in human development. Both anthropometric and dermatoglyphic measurements exhibit estimates of genetic determination of the order of 0.5 or higher [see for example Penrose (1954),Pons (1959),Osborn and de George (1959), Holt (1968). Loesch (1971), Ramesh and Murty (1977)],and both are often presented as classic examples of quantitative traits under the control of many loci. It is difficult to visualize the genes responsible for dermatoglyphic traits later influencing growth potential. Conversely, given the early fetal stage at which dermatoglyphic patterns are established, adult size can scarcely feed back into dermatoglyphic traits, with the exception of atd angle. But can there be a basis for an interaction between the two sets of measurements? Penrose (1965)postulated that the fact that the dermatoglyphic ridges run predominantly at right angles to the long axes of the limbs suggests that they take the shortest course possible during development, and that the symmetry of most triradii has the same implication. This led him to suggest that the permanent dermatoglyphic configuration is the result of laying a carpet of parellel lines, in some ways as economically as possible, over the contours presented by the fetal hand. In accordance with present findings we suggest that at the time that dermatoglyphic carpet is laid down, the responsible factors are interacting with other (genetic) factors ultimately related to body and hand shape. An indication 101 that these ‘other’factors find early expression is provided by the observation of Garn et al. (1975) that adult metacarpal-phalangeal proportions are attained as early as week 13, whereas the epidermal ridges do not complete their differentiation until week 16 (Cummins and Midlo, 1943).Further evidence for the kind of ‘prenatal communalities’ discussed here is provided by Babler’s (1980) observation that the thickness of the dermis is correlated with inter-ridge distance. ACKNOWLEDGMENTS The cooperation of the Center for Growth and Development of the Chilean National Health Service and the support of Project B 518 8135 of the Servicio de Desarrolo Cientifico, Artistic0 y de Cooperacion Internacional and Project PNUDKJNESCO RLA 781024 are gratefully acknowledged. LITERATURE CITED Babler, WJ (1980) Dermatoglyphics: Prenatal communalities in developing human hand. Am. J. Phys. Anthrop. 52202. Correa, N, Zambra, R, Gomez, G. Vargas, Y, Espinoza, V. Ortega, E, Pagliero, B. and Valdes, A (1976)Frecuencia de 10s grupos sanguineos ABO y Rh en donantes de sangre del Area Norte de Santiago. Santiago Rev. Tec. Med. 210-13. Cummins, H, and Midlo, C (1943)Finger Prints, Palms and Soles. Philadelphia: Blakistan Co. Garn, SM, Burdi, A, Bahler, W, and Stinson, S (1975)Early prenatal attainment of adult metacarpal-phalangeal rankings and proportions. Am. J. Phys. Anthrop. 43327-332. Holt. S (1968) The Genetics of Dermal Ridges. Springfield, Illinois: C. C. Thomas. Loesch, D 11971) Genetics of dermatoglyphic patterns of palms. Ann. Hum. Genet. 34277-293. Mood, AM, and Graybill, FA (1963) Introduction to the Theory of Statistics. New York: McGraw-Hill. Osborn, RH. and de George, FV (1959)Genetic Basis of Morphological Variation. Cambridge: Harvard University Press. Penrose, L (1954) The distal triradius on the hands of parents and sibs of mongo1 imbeciles. Ann. Hum. Genet. 19:lO-27. Penrose, L (1965) Dermatoglyphic topology. Nature 205:544-546. Pons, J (1959) Quantitative genetics of palmar dermatoglyphics. Am. J. Hum. Genet. 11252-256. Ramesh. A, and Murty. IS (1977)Variation and inheritance of relative length of index finger in man. Ann. Hum. Biol. 4:479-484. Tillner. I (1942) Untersuchungen uber Papillamuster. inbesondere im Hinhlick auf den Korporbautypus. Zeitsch. fur Menschl. Vererbungs. u. Konst. 26:93-128. Weiner, JS, and Laurie, J A (1969) Human Biology-A Guide to Field Methods. I B P Handbook # 4 9 . Oxford Blackwell Scientific Publications. Wendt, GC 11954) Konstitutiou und Fingerleisten. Z. Menschl. Vererb. u. Konstit. lehre. 32:116-125. Wichmann. D (1963) Uber nechtlineare Korrelation der quantitativen Werte der Fingerbeerenmuster mit der Konper-Lohe. Anthrop. Anz. 26:55-61. Valenzuela, CY, Rothhammer, F, and Chakraborty, R (1978) Sex dimorphism in adult stature in four Chilean populations. Ann, Hum. Biol6;533-538.