MOST persons prefer to use their right hand (RH) for one-handed tasks, such as writing, throwing a ball, hammering, sawing, cutting with a knife, and so on, but a significant minority prefer to use the left hand (LH) or either hand (ambidextrous). The term NRH (for non-right hander) is used here to represent both left-hand and ambidextrous users. The question of whether hand-use preference, referred to as handedness, is specified by biology/genetics, by cultural training, by pathological reasons such as birth stress, or by a combination of these etiologies has been addressed for decades and the debate continues. Several observations suggest problems with a strict genetic etiology. For example, not all the children of RH × RH are RH; not all the children of NRH × NRH are NRH; 18% of monozygotic twins are discordant for handedness even though co-twins possess identical genetic makeup; and cultural pressures can change some persons' preferences (Rife 1940). A relatively recent “random-recessive model” postulated, first, that a single hypothetical gene named RGHT1 (for specifying right handedness) is fully penetrant, causing persons with one or two RGHT1 copies to become RH; second, that the recessive nonfunctional allele r/r (r for random) homozygous persons develop as a 50:50 mixture of RH and NRH; and third, that monozygotic twins discordance results from randomness due to their r/r genotype (Klar 1996).

For etiology determination, one of the best definitions of handedness was proposed by Rife (1940). This rather strict definition designates a person as RH if the right hand is preferentially used for all 10 tasks such as to throw a ball, use a spoon, saw, sew, shoot marbles, bowl, cut with a knife, cut with scissors, hammer, and write. NRH are defined as those who do any number of these tasks with their left or with either hand. Following this definition, the random-recessive model was supported by several findings: the value of 8.0% NRH children born to RH × RH parents (Rife 1940) is as predicted (Klar 1996); a three-generational study suggested that the RH progeny of a LH × LH cross possess the r/r genotype (Klar 1996); the study of discordant twin progeny suggested that such twins possess the r/r constitution (Klar 2003). Furthermore, a recent study noted an interesting association of handedness with the direction of parietal (scalp) hair-whorl rotation found on the top of the human head (Figure 1). Specifically, it was discovered that NRH comprise a random mixture of clockwise (C) and counterclockwise (CC) persons and that persons chosen for their CC orientation were evenly divided between RH and NRH. In contrast, most people (>90%) in the general public are RH and most persons (>90%) support a C hair-whorl rotation. On the basis of such partial association of seemingly unrelated traits, it was suggested that both handedness and hair-whorl orientation are controlled by two alleles of a single gene. Individuals with the RGHT1 allele are RH and develop C hair whorls, and in r/r individuals there is no association between the traits, which are randomly distributed to the left vs. right side of a person (Klar 2003). However, conclusions of the hair-whorl study were challenged in a recent commentary. In the commentary it was suggested that most NRH result instead from a combination of causes, most from damage to the language-processing left brain hemisphere owing to birth stress (“pathological left-handers”) and a minority from learned reasons (“learned left-handers”), and the authors speculated that at most <20% of NRH may be genetically determined (Ehrman and Perelle 2004).

• Download figure
• Open in new tab
• Download powerpoint

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.—

The parietal hair-whorl phenotype. Most persons have a single hair whorl (top) and the minority exhibits two whorls (bottom). Among those with single whorls, the minority exhibits counterclockwise (top) rotation and the majority develops the clockwise orientation (not shown). Thus, the majority of people have a single clockwise hair whorl (Klar 2003).

A major problem faced in studies of a complex trait such as handedness concerns the criteria used to define a person's handedness. It is likely that, by varying the definition of a complex trait or the population ethnicity under study, results of different studies could vary significantly, making it difficult to compare conclusions of one study with those of others. Fortunately, a common human trait that is immune to subjective errors in scoring the phenotype exists. In >95% of humans, there is a single hair whorl on the scalp, and in the rest of humans there are two whorls (Figure 1); more than two whorls are extremely rare. Also, the whorl orientation does not change with aging or by the direction in which the hair is combed. Of equal importance, the phenotype is not subject to cultural influences or to birth stress. Thus, the hair-whorl orientation is a much more definitive phenotype as compared with handedness, and it should be exploited for deciphering the etiology of handedness in the general public. Therefore, to scrutinize the genetic interpretation for partial correlation between the traits of handedness and hair-whorl orientation proposed in the random-recessive model (Klar 2003) and to address the validity of the birth stress etiology (Ehrman and Perelle 2004), family studies concerning inheritance of the hair-whorl trait would be very informative for deciphering the handedness etiology of the general public. Fortunately, such an extensive and highly informative study was published long ago in 1927!

The most telling result concerns the study of inheritance of the hair-whorl trait in nuclear families (Schwarzburg 1927). The single gene, two-allele model was advanced whereby individuals containing one or two copies of the dominant allele (designated R) develop C whorls and all the recessive nonfunctional persons homozygous for the r/r alleles develop CC whorls (Schwarzburg model, Table 1). The study concluded that a single gene controls hair-whorl orientation, clockwise as a dominant trait; that an individual's sex does not influence hair-whorl orientation; and that hair-whorl orientation is not linked to eye color. According to the model, the frequency of 0.286 CC progeny found in C × CC families suggested 0.286 r:0.714 R allele ratios in C parents. According to this derived allele frequency, C × C families should produce 0.286 × 0.286 = 0.081 fraction of children supporting CC orientation. However, the predicted 0.081 value differed significantly from the observed 0.158 value (χ² = 6.45, d.f. = 1, P < 0.05), a result that contradicted the model (Table 1). We propose that the contradiction derives from an incorrect feature of the model that required all r/r individuals to develop CC orientation.

According to the random-recessive model, however, r/r individuals should exhibit 50% C:50% CC whorl orientation ratio (Klar 2003). Consequently, the frequency of 0.286 CC children born to C × CC parents predicts that the 0.286 × 2 = 0.572 fraction of the progeny should possess the r/r genotype. Therefore, their C parents should harbor 0.572 r:0.428 RGHT1 allele frequencies. According to this allele frequency, the C × C families are predicted to produce 0.572 × 0.572 × 50% (50% for the randomness parameter of the random-recessive model) = 0.163 CC progeny frequency (Table 1). The predicted value is remarkably similar to the observed value of 0.158 (χ² = 0.01, d.f. =1, P > 0.90). In the study concerning the CC × CC cross there was only one family in which all three children exhibited CC orientation (Table 1). The small number of three persons is statistically too small to differentiate between the models under consideration, and the result is not statistically inconsistent with the random-whorl orientation expectation of the random-recessive model.

A recent study (Klar 2003) suggested that the RGHT1 gene functions for brain hemisphere left-right laterality development so that the left-brain hemisphere processes language and that the gene functions to couple the dominant hemisphere to the development of both right-hand preference and clockwise hair-whorl rotation. It follows that r/r individuals develop these traits, but at least the traits of handedness and hair-whorl rotation are randomly distributed with respect to each other and to the left-right axis of a person (reviewed in Klar 2004a). The etiology of handedness continues to be debated despite hundreds of studies conducted over many decades. The hair-whorl study was performed a long time ago. By advancing an alternate explanation to the results, rediscovery and significance of the study to decipher the handedness etiology is recognized here. Ironically, and from a historical perspective, it is noteworthy that the hair-whorl trait was one of a handful of very early human traits exploited to study genetics in humans, the others being eye color, blood groups, the M and N agglutinogens, presence of hair on the second joint of the digits, and taste deficiency. The 1927 hair-whorl study provides support to the random-recessive model originally advanced to explain the etiology of handedness. Here I conclude that the same genetic mechanism explains the inheritance of the hair-whorl phenotype in the general German population. Thus the suggestion that handedness and hair-whorl traits develop from variation in a single gene/two-allele hypothesis (Klar 2003) is supported by the 1927 study. As the nonalterable whorl orientation is established prior to birth, the finding of association between handedness and the hair-whorl traits, combined with other similar arguments (Klar 2004d), argues against birth stress or other learning-based etiologies (Ehrman and Perelle 2004) for handedness in the general public.

A very significant variation in the handedness allele frequency is noteworthy in these two studies. In single-whorled persons only 8.4% (n = 500) were reported to be CC in the U.S. population (Klar 2003), while nearly 20% (n = 3960) exhibited CC rotation in the German population (Schwarzburg 1927). The r allele frequency derived by Schwarzburg should be doubled according to the random-recessive model. This difference suggests a significant variation in the allele frequencies of the two populations, results of which were recorded over seven decades apart. Further studies are needed to quantify variations in the handedness allele frequencies in other geographical regions, ethnic populations, and even people from different professions by exploiting the easily discernible hair-whorl phenotype. It should be noted in this context that LH is rare among African children (Verhaegen and Ntumba 1964), while in the United States, Jewish college students were reported to be twice as likely to be NRH compared to the non-Jewish population (Rife and Schonfeld 1944). In addition, the hair-whorl phenotype may be exploited to study the contribution of genetics concerning development of brain hemispheric laterality to other human behaviors, including a person's life style. For example, it was recently reported that samples of homosexual men exhibited a 3.6-fold excess of CC orientation as compared to men at large in the U.S. population (Klar 2004b). With overwhelming experimental support garnished for genetic etiology, researchers should next address what biological function the RGHT1 gene performs to influence left-right laterality development of neuronal tissues influencing brain laterality, handedness, and hair-whorl orientation. At present only a single suggestion has been made in this context. On the basis of strictly genetic results, it was recently suggested that the RGHT1 gene facilitates patterned segregation of chromosome 11 chains in mitosis to differentiate sister cells at a specific cell division in embryogenesis, ultimately leading to brain hemispheric specialization (Klar 2002, 2004c).

• Received December 16, 2004.
• Accepted April 25, 2005.

• Genetics Society of America