One hundred paraffin-embedded cervical biopsy specimens were tested for the presence of human papilloma virus (HPV) by in situ hybridization (ISH), and by direct and indirect in situ PCR (IS-PCR) in order to evaluate the efficiency of the different in situ methods in detecting HPV infection. ISH was performed using either commercial DNA probes or a cocktail of 5'-digoxigenin labeled oligoprimers. The same were used for ISH during indirect IS-PCR. To enhance the sensitivity of ISH several polymers, i.e., polyvinyl alcohol (PVA), polyethylene glycol, and polyvinylpyrrolidone were added to the alkaline phosphatase nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP) reaction. Furthermore, tyramide signal amplification (TSA) was tried for signal amplification. Those samples treated with PVA during the NBT/BCIP reaction did not show any signal amplification whereas those treated with TSA exhibited a dramatic increase in sensitivity with usually acceptable signal to noise ratios. Our results show that, regarding sensitivity, ISH with subsequent signal amplification by TSA can be used as an almost equivalent alternative to the more cumbersome IS-PCR on routinely processed tissue specimens. When considering reproducibility, it is superior to IS-PCR.