Abstract: Formalin-fixed specimens were of profound importance to the study of evolution, systematic, phylogeny and genetic structure of populations. However, the PCR analysis of DNA isolated from formalin-fixed tissue could be difficult because of the inherently poor quality of template DNA available for amplification. As a consequence of fixation process, DNA was complex with proteins and is often nicked, giving relatively short fragments. Such samples were often of low DNA concentration and poor quality. Formalin-fixation had a profound effect on the molecular arrangement of nucleic acids in the tissue and resulted almost invariably in DNA degradation of various degrees. Extraction, purification and amplification of DNA from formalin-fixed tissues were influenced by a multitude of factors. In this report, we introduced a method of DNA extraction and PCR amplification of relatively large target fragments from formalin-fixed, fluid-preserved tissues. Main modifications include： long-time preservation in buffer, short-time heating, and drying in a vacuum centrifuge during pre-disposal; more proteinase K and sodium SDS during extraction; performing PCR immediately after digestion, prolonging annealing time, and increasing the number of cycles. The feasibility of this method was tested by PCR amplification. The nucleotide DNA of cloned products from formalin-fixed specimens were sequenced and the results were compared with the sequences obtained from fresh and ethanol-fixed tissue and published data. Comparison of the DNA sequence data from the formalin-fixed tissues with that from the frozen and ethanol-fixed tissues demonstrated this method is reliable. We also found that the pre-fixation time could be an important factor determining the quality of DNA.