Saturday, November 10, 2012

Are women changing color?

Tanned arm (source). Underarm skin color used to be a reliable measure of constitutive pigmentation (skin color before tanning). Is it still?

After puberty, girls become lighter-skinned than boys. This sexual differentiation has been shown to be hormonal in origin by a digit ratio study (Manning, Bundred, and Mather, 2004), by studies on normal, castrated, and ovariectomized individuals (Edwards & Duntley, 1939, Edwards & Duntley, 1949; Edwards et al., 1941), and by a twin study of boys and girls at different stages of puberty (Omoto, 1965). In addition, women are lighter-skinned than men in a wide range of human populations, although the sex difference is smaller in very light-skinned and very dark-skinned populations (Frost, 2007; Madrigal & Kelly, 2006).

This sex differences is reversed, however, in a recent study of subjects from Ireland, Poland, Italy, and Portugal: 

Surprisingly, we find in our cohort that males have lighter skin pigmentation (lower M) than females in all four European countries. (Candille et al., 2012)

All of the subjects were post-puberty (largely in their early to mid-20s) and the measurements came from the upper inner arm, a site relatively unaffected by tanning. So what happened?

First, a different kind of instrument was used to measure skin color:

Our results in populations of European ancestry contradict earlier anthropological studies that have concluded females are more lightly pigmented than males in most populations (reviewed in [2]). One potential reason for the conflicting results is the different instruments used. In early studies, which used the Evans Electric Limited (EEL) and Photovolt broad-spectrum spectrophotometers, skin pigmentation estimates may be confounded by the hemoglobin level to a greater extent than for the DermaSpectrometer used in the present study (Candille etal., 2012)

The Candille et al. research team didn’t measure skin color at all wavelengths of visible light. They focused on those wavelengths that melanin absorbs. Previous researchers had studied the overall visual difference between male and female skin, which is due as much to differences in hemoglobin as to differences in melanin. In short, men are browner and ruddier than women.

It is misleading, then, to state that the male subjects had lighter skin pigmentation. Hemoglobin too is a skin pigment.

Nonetheless, it’s still surprising that the women had more melanin than did the men, and this was on the upper inner arm—a body site relatively unexposed to tanning. That finding does contradict earlier studies. According to the earliest major one on human skin color:

It is generally known that women are lighter colored than men. From our studies it is apparent that this is due to the female skin containing less blood and melanin (Edwards & Duntley, 1939)

The study also found that this sex difference was smaller over much of the body surface, apparently because women exposed more of it to the sun:

The skin of women is generally poorer in melanin than that of men (fig. 16). However, because of their manner of dress, women of the white races show a comparatively higher pigmentation of the shoulders, upper chest and upper extremities (Edwards & Duntley, 1930)

For this reason, most subsequent researchers have measured skin color at the upper inner arm, this site being relatively unexposed to the sun and thus providing a better measure of constitutive pigmentation.

The study’s authors, Edwards and Duntley, went on to study castrated and ovariectomized subjects to understand how the sex hormones affect skin color. Although testosterone had a stronger impact than did estrogen on skin pigments (both melanin and hemoglobin), the absence of these hormones did not eliminate all of the sex-specific characteristics of male and female skin. The two researchers concluded:

Our observations show that ovariectomy does not entirely change the basic peculiarities which distinguish the female from the male skin. Similarly, the male castrates previously studied had maintained to some extent, their differences from the female. In both sexes, therefore, we can assert that sex differences are basically due to genic influence. This base line of pigmentary characteristics is then modified by the presence of the sex hormones. Unlike the situation in many other animals, where reactivity to the hormones is localized in special areas, the human skin reacts probably in its entirety (Edwards & Duntley, 1949)

This “genic influence” is now known to be the prenatal hormonal surge that determines whether an individual will develop as a boy or a girl. “Presence of the sex hormones” would be better described as “circulating sex hormones”—the individual’s current hormonal status. This is the conclusion of Manning, Bundred, & Mather (2004) in their digit ratio study:

We find that 2D:4D and female ‘constitutive’ pigment scores are negatively related. Since women with light skin tend to have high ‘feminised’ digit ratios, and there is evidence that the 2D:4D ratio is positively related to prenatal oestrogen levels, it seems that oestrogen may have an early organisational effect on skin pigment in women. The absence of a relationship between 2D:4D and skin colour in men suggests that other factors, such as prenatal testosterone, may obscure the in utero effects of oestrogen.

Other studies have focused on the sexual differentiation of skin color at puberty. All of them conclude that female skin loses melanin at puberty:

Though both the boys and the girls of the two populations show a decline in the melanin content of the skin when passing through adolescence, yet the decline in the pigment is so much pronounced in the girls that it reverses the sex differences in skin pigmentation. The boys, who were lighter in skin colour than the girls at younger ages (i.e. below 13 years), become darker than the girls at advanced age (at least up to 16 years) (Kalla, 1973).

In the above study, skin color was measured on the upper inner arm at a time of year (December-January) when tanning was minimal. Kalla (1973) concluded that the external environment could not have caused this sexual differentiation, the only possible explanation being “the puberty changes of the endocrine status.”

This was likewise the conclusion of Mesa (1983):

[…] the remarkable changes in pigmentation that occurred during the pubertal period (and which are related to hormonal changes) are a major factor in the differences (between the sexes and between the [upper inner] arm and the forehead) that were found at older ages and, also, in the differences observed in the adult population.

So how can we explain the recent findings by Candille et al. (2012)? Perhaps underarm skin color is no longer a reliable measure of constitutive pigmentation, i.e., color of untanned skin. Removal of underarm hair has become the norm in recent years, especially among European women. This has led to concern about tan lines in the underarm region and a consequent desire for “full body tanning,” as described below by one e-columnist:

How to Tan Underarms in a Tanning Bed

By Lauren Wise, eHow Contributor

A tanning bed provides a quick, easy, natural looking tan to a man or woman's body. The bed nearly triples your tanning time with the extra strong rays it provides. Unfortunately, since it is difficult to move around in a tanning bed, it is hard to tan every angle and crevice of the body, in particular the underarms. It is easy to change this by alternating your movements within the tanning bed throughout the session.


Candille, S.I., Absher, D.M., Beleza, S., Bauchet, M., McEvoy, B., et al. (2012). Genome-Wide Association Studies of Quantitatively Measured Skin, Hair, and Eye Pigmentation in Four European Populations. PLoS ONE 7(10): e48294. doi:10.1371/journal.pone.0048294

Edwards, E.A. & S.Q. Duntley. (1949). Cutaneous vascular changes in women in reference to the menstrual cycle and ovariectomy, American Journal of Obstetrics and Gynecology, 57, 501-509.

Edwards, E.A. & S.Q. Duntley. (1939). The pigments and color of living human skin , American Journal of Anatomy, 65, 1-33.

Edwards, E.A., J.B. Hamilton, S.Q. Duntley, & G. Hubert. (1941). Cutaneous vascular and pigmentary changes in castrate and eunuchoid men, Endocrinology, 28, 119-128.

File:Skin tanning.JPG

Frost, P. (2007). Comment on Human skin-color sexual dimorphism: A test of the sexual selection hypothesis, American Journal of Physical Anthropology, 133, 779-781.

Kalla, A.K. (1973). Ageing and sex differences in human pigmentation, Zeitschrift für Morphologie und Anthropologie, 65, 29-33.

Madrigal, L. & W. Kelly. (2006). Human skin-color sexual dimorphism: A test of the sexual selection hypothesis, American Journal of Physical Anthropology, 132, 470-482.

Manning, J.T., P.E. Bundred, & F.M. Mather. (2004). Second to fourth digit ratio, sexual selection, and skin colour, Evolution and Human Behavior, 25, 38-50.

Mesa, M.S. (1983). Analyse de la variabilité de la pigmentation de la peau durant la croissance, Bulletin et mémoires de la Société d'Anthropologie de Paris, t. 10 série 13, 49-60.

Omoto, K. (1965). Measurements of skin reflectance in a Japanese twin sample, Journal of the Anthropological Society of Nippon (Jinruigaku Zassi), 73, 115-122.

Wise, L. (2012). How to Tan Underarms in a Tanning Bed, eHow style


Sean said...

Well if men have less melanin in their skin than women that would be more of a problem for the simple UVB-latitude-melanin hypothesis proponded by Nina G. Jablonski than sexual selection.

Bilirubin is also important in the skin color difference between the sexes, and men have more.
Comprehensive candidate gene study highlights UGT1A and BNC2 as new genes determining continuous skin color variation in European. UGT1A polymorphisms are related to bilirubin plasma levels, but there is a gender-associated difference

Sean said...

The wonders 'natural selection' works on skin color here. So fair skin evolved to give you cancer even if you don't go out in the sun, and lower vitamin D levels if you go out in the sun.

Anonymous said...

To some extent, isn't the whether lighter female skin colour is environmental or genetic/hormonal in origin kind of irrelevant to sexual selection?

If women had lighter skin historically, across the evolutionary history (due to lesser exposure - e.g. boys got tans and girls didn't, due to men being out in the sun a lot), then that could still be the subject of selection by males...

It's true that if it is genetic / hormonal in origin, then it is more plausible to suggest it as true across the EEA / selective environment, but it does not render sexual selection implausible.

Sean said...

Anon, Nina G. Jablonski says the female need for rebuilding strong female skeletons between pregnancies may be the origin of skin color differences between the sexes. If white women have more melanin in their skin than men Jablonski's suggestion wouldn't make sense.

Even if the recent norm of Western women underarm shaving/tanning is not responsible for the study's finding that skin color due to melanin is darker in women, it remains true that men's skin is darker, due to it containing more blood, and in whites that difference happens to be salient.

Selection for dark male skin due to melanin would have been relaxed if there was strong sexual selection for lighter skin on women. It's known that Middle East populations have far more of a sex difference in skin color.

Anonymous said...

Anon, Nina G. Jablonski says the female need for rebuilding strong female skeletons between pregnancies may be the origin of skin color differences between the sexes. If white women have more melanin in their skin than men Jablonski's suggestion wouldn't make sense.

I'm not so sure what that has to do with my point, exactly. I thought you didn't buy that skin colour had any links to Vit D though? Are you saying that a darker skin in females would be disconfirmatory for the Vit D hypothesis for light skin evolution.

The suggestion by Jablonski you have highlighted does seem arbitary though - why would not men with proportionately larger skeletons benefit more from Vit D?

Regarding the other point, male skin which is darker because it is redder leading to selection for female skin which is lighter because it is less brown seems more of a difficult and less tenable phenomenon than if it were male skin which is less brown.... (in fact, if differences in skin redness between men and women were more salient than brownness per se, it would seem to push to sexual selection being more likely in East Asia, where skin is much red, but not that much more brown...).

Sean said...

As far as I know women have much the same amount of hemoglobin the world over. The study was measuring melanin only, and it's melanin that is supposed to vary between populations by latitude as a result of the need for vitamin D synthesis. Women supply calcium to the fetus by giving it their own body calcium. Nina G. Jablonski suggests they have to replenish their stores quickly between pregnancies to give birth repeatedly, and that is why women have less melanin in their skin than men. But if women don't have more melanin Nina's hypothesis has a problem.

Sean said...

Anon, Even if you stay in the shade it is virtually imposible to avoid UVB hitting exposed skin; UVB photons ricochet off the ground, clouds ect to hit the skin from all angles.

The Dietary Reference Intakes recomendations for 'vitamin' D assume no skin synthesis of vitamin D whatever. A totally unrealistic assumption.

Thinking about tanning a little more, the parts of the body that never get sun wouldn't be seen at all and won't be involved in sexual selection between prospective mates. Maybe a lesser propensity to darken from UV where the skin is visible to a prospective mate, such as the face, is the focus of the force of selection on womens skin, but where the skin is rarely exposed to UV or the eyes of a prospective mate pre mating (eg unshaved inner arm) there is little or no difference.