⇐ Asparagus urine |
Beeturia ⇒
Table of Contents
Genetics of earlobes
Summary
You may have heard that there are two kinds of earlobes, "attached" and "free," with a free earlobes being a dominant genetic trait. Neither part of that is true! Earlobes actually show a smooth range of variation from free to attached, with many intermediates, and they are affected by many genes, not just one. Don't worry if your family includes two parents with attached earlobes and a child with free earlobes. And if you're a teacher, don't use earlobes as an example of a simple genetic trait.
Are there just two kinds of earlobes?
Lots of biology teachers use earlobes as an example of a human genetic trait, and they divide them into two kinds, free (F) and attached (A). "Free" earlobes have the point where the earlobe joins the head up near the base of the ear, well above the lowest point of the earlobe, while "attached" earlobes have the attachment point further down the head. However, since the early days of earlobe genetics research, scientists have pointed out that there are many people with intermediate earlobes (Quelprud 1934, Dutta and Ganguly 1965). Weiner (1937) said that "arbitrary classification into two sharply defined types...gives a false picture, since all gradations between the two extremes are encountered." Instead of dividing earlobes into two categories, some researchers have divided them into three (Sharma et al. 2007, El Kollali 2009), four (Weiner 1937), or six (Beckman et al. 1960) categories. No one has actually measured the attachment points and plotted the distribution on a graph, but I'm confident that it would show a smooth distribution with lots of intermediates, and dividing earlobes into any number of distinct categories is completely arbitrary.
To illustrate this, I searched for pictures of professional bicyclists (because they mostly have short hair, and because I'm a fan of bicycle racing), found 12 with their ears showing, and arranged them from free to attached. It's pretty clear that there aren't just two kinds of earlobes.
When I was a graduate student, working as a teaching assistant for an introductory biology lab, one lab had the students filling out worksheets that asked whether they have free or attached earlobes, could roll their tongues, and many of the other traits that I discuss in these pages. I took a good look at my own ears, to see whether I could use myself to illustrate "free" or "attached," and realized that my own earlobes are exactly halfway in between the two extremes. Once I realized that the lab manual's information about earlobes was oversimplified, I started to wonder about the other traits used to teach human genetics, which eventually led to me writing these pages many years later.
Family studies
Carrière (1922) looked at 15 families and concluded that attached earlobes (A) were dominant. However, all of the offspring of A×A matings had attached earlobes, and there were no F×F matings, so his data are consistent with either free or attached being dominant.
Powell and Whitney (1937) looked at one family and concluded that attached earlobes were recessive. Wiener (1937) divided earlobes into four arbitrary groups, from 0 (completely free) to 3 (completely attached). All possible matings, from 0×0 to 3×3, produced some intermediate earlobes. Wiener (1937) concluded that earlobes were determined by more than one gene, or by a single gene with more than two alleles.
Lai and Walsh (1966) called earlobes in which the lowest point on the earlobe was the attachment point "attached," and they classified all other earlobes as "free." They recorded the following data on families in New Guinea:
| Parents | F offspring | A offspring | Percent F |
|---|---|---|---|
| F×F | 12 | 22 | 35% |
| F×A | 72 | 114 | 39% |
| A×A | 37 | 90 | 29% |
If attached earlobes really were caused by a recessive allele, two parents with attached earlobes could not have a child with a free earlobe. There are slightly more A offspring from A×A matings, but the large numbers of F offspring from A×A matings and A offspring from F×F matings indicate that this is not a one-locus, two-allele trait.
Mohanraju and Mukherjee (1973) performed a similar study in India and found similar results:
| Parents | F offspring | A offspring | Percent F |
|---|---|---|---|
| F×F | 13 | 1 | 93% |
| F×A | 7 | 7 | 50% |
| A×A | 5 | 29 | 15% |
They found a much stronger association between parents and offspring, but the five F offspring of A×A matings mean that this is not a one-locus, two-allele trait.
Genome studies
There have been three genome-wide association studies of earlobe attachment, in which hundreds of thousands of single-nucleotide polymorphisms (SNPs) are surveyed in thousands of people. Eriksson et al. (2010) looked at 3915 customers of 23andMe, all of northern European descent. They answered an online question, "Are your earlobes attached or unattached." Eriksson et al. (2010) did not find any SNPs with a significant relationship to earlobe attachment.
Adhikari et al. (2015) photographed the ears of 5000 people from Latin America and divided them into three categories, attached, intermediate, and unattached. They found statistically significant associations of earlobe attachment with two genes, ectodysplasin A receptor (EDAR) and specificity protein 5 (SP5), both of which are known to be involved in development and to affect variation in other morphological traits.
Schaffer et al. (2017) photographed the ears of 1791 European-Americans, 5062 Latin Americans, and 2857 Chinese, and divided the ears into three categories. They also had a sample of 64,950 23andMe customers of European ancestry who self-reported whether their earlobes were attached or free. They identified 49 regions, distributed across 15 chromosomes, with a statistically significant association to earlobe attachment. Many of the regions included genes previously known to be involved in development or expressed in ear tissue, including EDAR and SP5.
Because the Adhikari et al. (2015) and Schaffer et al (2017) studies have shown that multiple genes affect earlobe attachment, it is clear that earlobe attachment is not a simple one-gene trait.
Further research
Any further genetic research on earlobe attachment would require genome-wide association studies, followed by sophisticated molecular and developmental investigation of the individual genes found to be involved. Before doing that, it would be useful if someone invented an objective, repeatable method for measuring the amount of earlobe attachment, instead of just dividing earlobes into two arbitrary categories. The measurement might be as simple as the distance between the earlobe attachment point and some point on the outer ear, such as the lowest point inside the shell of the outer ear, so it could be a fun classroom project.
References
Adhikari, K., et al. (38 coauthors). 2015. A genome-wide association study identifies multiple loci for variation in human ear morphology. Nature Communications 6: 7500.
Beckman, L., J.A. Böök, and E. Lander. 1960. An evaluation of some anthropological traits used in paternity tests. Hereditas 46: 543-569.
Carrière, R. 1922. Über erbliche Orhformen, insbesondere das angewachsene Ohrläppchen. Zeitschrift für Induktive Abstammungs- und Vererbungslehre 28: 288-242.
Dutta, P., and P. Ganguly. 1965. Further observations on ear lobe attachment. Acta Genetica 15: 77-86.
El Kollali, R. 2009. Earlobe morphology: a simple classification of normal earlobes. Journal of Plastic, Reconstructive and Aesthetic Surgery 62: 277-280.
Eriksson, N., J.M. Macpherson, J. Tung, L. Hon, B. Naughton, S. Saxonov, L. Avey, A. Wojcicki, I. Pe'er, and J. Mountain. 2010. Web-based, participant-driven studies yield novel genetic associations for common traits. PLoS Genetics 2010. 6:e1000993.
Lai, L.Y.C., and R.J. Walsh. 1966. Observations on ear lobe types. Acta Genetica 16: 250-257.
Mohanraju, C., and D.P. Mukherjee. 1973. Ear lobe attachment in an Andhra village and other parts of India. Human Heredity 23: 288-297.
Powell, E.F., and D.D. Whitney. 1937. Ear lobe inheritance: an unusual three-generation photographic pedigree chart. Journal of Heredity 28: 184-186.
Quelprud, T. 1934. Familienforschungen über Merkmale des äusseren Ohres. Zeitschrift fü Induktive Abstammungs- und Vererbungslehre 67: 296-299.
Shaffer, J.R., et al. (32 coauthors). 2015. Multiethnic GWAS reveals polygenic architecture of earlobe attachment. American Journal of Human Genetics 101: 913–924.
Sharma, A., N.K. Sidhu, M.K. Sharma, K. Kapoor, and B. Singh. 2007. Morphometric study of ear lobule in northwest Indian male subjects. Anatomical Science International 82: 98–104.
Wiener, A.S. 1937. Complications in ear genetics. Journal of Heredity 28: 425-426.
Go to John McDonald's home page
This page was last revised October 26, 2026. ©2026 by John H. McDonald. Content is available under the Creative Commons Attribution-Noncommercial license.