Penetration studies are commonly performed on in vitro models, presumably due to a lack of non-invasive in vivo methods. To date, it is not clear whether in vitro models are suitable to reflect the in vivo conditions for percutaneous penetration. Apart from inter and intraspecies skin differences, the excision of a skin sample may influence the penetration rate inter alia as a result of the contraction of the elastic fibres in the skin during excision. Therefore, the aim of the present study was to investigate the follicular reservoir of the hair follicles of human skin in vivo and in vitro utilizing the method of differential stripping. The results obtained revealed a significantly reduced follicular reservoir in vitro, which was only 9.5+/-10.6% of the in vivo reservoir. These results are important for the interpretation of earlier and future penetration investigations. It can thus be assumed that excised skin models are suitable for penetration studies only to a limited extent, as follicular penetration is greatly diminished due to the contraction of the elastic fibres of the skin.