48 JOURNAL OF COSMETIC SCIENCE and loose. Washing the hair subsequent to UV light irradiation causes some melanin granules to effuse from the hair, which in turn leads to lightening. The ratio of eumelanin to pheomelanin has been quantified by some researchers. The results always agree that eumelanin is predominant in black or dark brown hair. In the case of blond or red hair, however, the results are sometimes inconsistent with each other. Some researchers reported that eumelanin prevails even in blond and red hair (2,8). Contrary to this, some researchers have reported pheomelanin to be predominant in red hair (9). Others mention that pheomelanin is predominant in blond hair (10). This discrepancy must come from not only the different measuring methods, but also from the hair samples used. In this study we have examined the photolightening behavior of red hair, whose melanin is predominantly pheomelanin. The blond hair used in our previous study (1) contained both eumelanin and pheomelanin. From this, the photo­ sensitivity of each type of melanin can be clarified by comparing photolightening behaviors between red hair and blond hair. There have been many studies on the photosensitivity of the above-mentioned two types of melanin, and two conflicting views exist. Some researchers have argued that phe­ omelanin is more easily degraded by UV light than eumelanin (11�13), often grounding their argument in the fact that redheads are highly susceptible to skin cancer. They consider that melanin in the skin of redheads is mainly pheomelanin, which is easily decomposed by UV light, and therefore cannot exhibit the role of protecting the skin against sunlight. On the contrary, other researchers offered a counterargument that pheomelanin was rather more photoresistant than eumelanin, based on the fact that the color of red hair does not change very much during summer (14). They noted that most of the other studies were undertaken using synthetic or isolated melanin, clarifying that it was more important to investigate the photosensitivity of melanin in natural condi­ tions. The controversy surrounding the photosensitivity of the two types of melanin has not been clarified for some decades. It is considered that synthesized melanin probably shows different properties from natural melanin, with isolated melanin granules denaturalized during the isolation process. On the other hand, irradiation of hair fibers can provide valuable information about the photochange of intact melanin in nature. However, the relationship between the degree of photodegradation of each type of melanin and the degree of color change of hair has not yet been established. Hence this can possibly lead to a misunderstanding if the color change in hair is related to the level of photodegradation of melanin. In our work, we examined chemically intact melanin granules of red hair using the same method of irradiating a cross section of hair that was applied to blond hair in the previous study (1). The photosensitivity was estimated by comparing the percentages of the disappeared melanin granules before and after irradiation. MATERIALS AND METHODS RED HAIR Chemically untreated Caucasian red hair was purchased from International Hair Im­ porters & Products Inc., New York. Tresses with a weight of 0.5 g and 8 cm in length

PHOTOLIGHTENING MECHANISM OF RED HAIR 49 were prepared. The tresses were washed three times with aqueous sodium laureth sulfate solution (15%), adjusted to pH 7 with phosphoric acid, and air dried. TREATMENT OF RED HAIR WITH PAP AIN Red hair (50 g) was treated with papain to obtain brown granules. The treatment method was performed according to Zahn's method for the isolation of melanin granules (15). Crude granules (500 mg) isolated from the 50 g of red hair were washed twice with a hexane/isopropanol/water (6/6/1) mixture (200 g), and further washed three times with deionized water (200 ml) to purify them prior to drying under reduced pressure (10 cmHg, at 60°C, for 1 hr). The SEM image of the obtained granules is shown in Figure la. For comparison, the SEM image of the eumelanin granules isolated in the same way from Chinese black hair is shown in Figure 1 b. The granules obtained from the red hair were mainly of indeterminate form, with sizes of around 0.4 µm, which coincides with published data for pheomelanin granules (16). A small amount of rice-like granules, which are considered to be eumelanin granules, was also observed. IRRADIATION EXPERIMENTS Irradiation experiments were carried out with a solar simulator and some optical filters, under the same conditions as previously reported (1). VIS-light (wavelength 370-780 nm) and UV-light (wavelength 280-400 nm) intensities were representative of those found in natural sunlight. In this work, for each light irradiation experiment, one irradiation unit corresponds to a single light exposure in sunlight for one hour. Ir­ radiation was performed under dry conditions, between 25° and 35°C, and at 20- 30% RH. LIGHT IRRADIATION ON RED HAIR Red hair tresses were irradiated with VIS or UV light in the two cases detailed below. For each condition, six tresses were taken and their data averaged. Case 1: Irradiation with repeated washing. The hair tresses were washed at specific irra­ diation times (5-20 irradiation units) with aqueous sodium laureth sulfate solution (15%), adjusted to pH7 with phosphoric acid, rinsed, and then air dried. Thereafter, the color was measured and irradiation was carried out. a b Figure 1. Isolated granules through papain treatment. a: From red hair. b: From black hair.