To determine the effect of different fixatives on the recovery and detection of mRNAs from archival histopathology specimens, biopsies of normal skin were fixed in neutral and alcohol-buffered formalin, acetone, Carnoy's fixative, methacarn, and Bouin's solution. Tissue was routinely processed, and sections were either mounted for in situ hybridization or deparaffinized for RNA extraction. Extracted mRNA was reverse-transcribed using random hexamers, and the resulting cDNA was amplified by the polymerase chain reaction using primers specific for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and beta-actin. Amplification products of both GAPDH and actin could be detected by gel electrophoresis from tissues processed in all fixatives except Bouin's. A parallel study of formalin-fixed, paraffin-embedded archival biopsies accessioned since 1990 gave similar results. Less abundant mRNAs, such as those encoding interleukin-11 or the T-cell receptor beta-chain, could be detected by Southern blotting and hybridization with labeled oligonucleotide probes or by cloning and sequencing. In situ hybridization studies using oligonucleotide probes were most successful with tissue fixed in formalin, including both the experimentally fixed tissues and the archival biopsy samples. Thus, mRNAs may be isolated from and localized in formalin-fixed, paraffin-embedded archival material. Because dermatopathology laboratory archives typically contain samples from a wide spectrum of diseases that can be accessed without Human Investigation Committee approval, these laboratories represent a logical starting point for studying gene regulation and expression in skin.