The chromatic characteristics of skin color arise from the interactions of light (primarily absorption and scattering) with the epidermis and the dermis. The primary light absorbers in skin are hemoglobin and melanin. Most of scattering is attributed to collagen fibers and in pigmented skin to melanosomes. Traditionally skin redness is considered to arise due to locally elevated concentrations of hemoglobin, whereas skin pigmentation is attributed to melanin. In this study we attempt to understand better the contributions of these chromophores to the perceived skin color using spectral analysis of skin color reactions induced by ultraviolet (UV) irradiation or pressure. In the first experiment 12 individuals with skin phototypes III-IV were irradiated on the back using a solar simulator with doses ranging from 0.7 to 3 MED. The skin reactions were evaluated on days 1, 7, 14, and 21 after irradiation. Evaluations included diffuse reflectance spectroscopy (DRS) and clinical assessment of the erythema and the pigment reaction. Apparent concentrations of melanin, oxy-, and deoxy-hemoglobin were calculated from the absorption spectra. In the second experiment the levels of deoxy-hemoglobin of the volar forearm of ten volunteers were selectively altered by either application of a pressure cuff or by topical application of 3% H(2)O(2). Changes in skin color appearance were documented by photography, colorimetry, and DRS. In the UV exposure experiment all reactions were dose dependent. Oxy-hemoglobin values increased to a maximum on day 1, correlating well with the clinical evaluation of erythema, and then decreased exponentially to base line. Melanin showed a significant increase on day 7 and remained relatively constant for the next 3 weeks, correlating well with the clinical evaluation of pigmentation (tanning). Deoxy-hemoglobin increased slightly on day 1 and remained elevated for the next 2 weeks. Thus, deoxy-hemoglobin correlated moderately with the clinical erythema scoring on day 1 only, while it contributes significantly to what is clinically perceived as skin tanning on days 7 and 14. Application of pressure below the diastolic level increased deoxy-hemoglobin concentration as measured by DRS. This increase corresponded to a decrease of a "pigmentation" parameter (based on the L(*)a(*)b(*) scale) in a similar fashion that has been documented for increases in melanin concentration. Topical H(2)O(2) application reduced deoxy-hemoglobin levels as measured by DRS. This reduction coincided kinetically with a visible skin blanching. Application of pressure or H(2)O(2) did not significantly alter the levels of oxy-hemoglobin or melanin. In this report we present compelling evidence that deoxy-hemoglobin significantly contributes to the skin color appearance. Blood pooling, expressed as increased deoxy-hemoglobin, can contribute to what is visually perceived as pigmentation. Furthermore, we present that measurement of its contribution to the skin color appearance can only be accomplished with DRS.