Tongue Rolling in the Classroom

Abstract

Tongue rolling, the ability to shape the tongue into a longitudinal trough, is a trait that is commonly used in the classroom as an example of human inheritance. This polymorphic trait is easily determined and its mode of inheritance is presumed to be simple. According to Sturtevant (1940), who first described the inheritance of tongue rolling, the ability to do so is due to a pair of alleles. The ability to roll one's tongue is dominant to the inability to do so. According to this hypothesis, children of tongue rollers can be found to be unable to roll their tongue, but children of nonrollers should always be non-rollers. This was found to be untrue in our classes. From the pedigrees of hundreds of students, we found a few tongueroller students whose parents were unable to roll their tongue and students whose parents were tongue rollers but whose own parents were non-rollers. Intrigued by these results we surveyed the literature that revealed that the ability of tongue rolling is far from being a simple Mendelian trait. Though Hsu (1948) and Garhes (1952) confirmed Sturtevant's original hypothesis, a few researchers had doubts. Matlock (1952) studying 33 pairs of identical twins observed seven pairs who were discordant. From these observations he had to conclude that tongue rolling is not entirely hereditary. Reedy, Szczes, and Downs (1971) compared 193 pairs of identical and fraternal twins. They concluded that the data were inconsistent with the hypothesis of chance fluctuations and that the ability to roll one's tongue was strongly influenced by hereditary factors. However, Martin (1975) from a comparable study that included 47 identical and fraternal twins did not find any statistical evidence for greater concordance in identical twins. He concluded that there was no evidence for a genetic basis for tongue rolling. We decided to pool the data of Matlock, Reedy, Szczes, and Downs and Martin (table 1). The Chi Square of the pooled data is 1.89. Because the probability is greater than 5 percent, we have to conclude that there is no evidence for a greater concordance in identical twins. This can be further confirmed by another statistical test, which estimates the differences between the proportions of concordance in the fraternal and identical twins by calculating the con-