Although in vitro skin absorption studies often detect small residues of applied test material in the epidermis/dermis, it is uncertain whether the residue is within the living skin. We studied the dermal absorption of a hair dye hydroxyanthraquinone-aminopropyl methyl morpholinium methosulphate (HAM) in human skin in vivo and in vitro. In vivo, skin (back and scalp) received 0.5% HAM in a commercial formulation at 20microg/cm2 After 0.5 or 48h, skin was tape stripped, followed by cyanoacrylate biopsies (CAB). Sebum from scalp sites was collected for 48h. In vitro, skin was treated with 20mg/cm2 dye for 0.5h, penetration determined after 24h. In vivo, at 0.5h, total recovery (back) was 0.67microg/cm2 (tape strips+CAB). Fluorescence microscopy showed HAM in the hair follicle openings (HFO). At 0.5h, scalp tape strips contained 1.80microg/cm2, HFO 0.82microg/cm2. At 48h, HFO contained 0.21microg/cm2, sebum 0.80microg/cm2. In vivo, skin residues were in the non-living skin and eliminated via desquamation and sebum secretion. In vitro, the SC contained 1.50microg/cm2, epidermis/dermis 0.86microg/cm2, receptor fluid<0.04microg/cm2, a total of 0.90microg/cm2 was considered to be bioavailable. In vitro epidermis/dermis residues were nearly identical to those located in non-living skin in vivo. In conclusion, in vitro percutaneous penetration studies may produce seemingly bioavailable material , which raises the need for a Threshold of Skin Absorption (TSA) addressing a negligible dermal absorption in order to avoid unnecessary in vivo toxicity studies on substances that produce no significant human systemic exposure.