Mycoplasma is a common contaminant of tissue culture samples. Infection is persistent, difficult to detect and diagnose, and very difficult to cure. The concentration of mycoplasma in infected cultures can be as high as 10(7) colony-forming units per mL, and their presence can change many of the cell reactions, including altering cell growth rate, inducing morphological changes or cell transformation, and mimicking virus infection. Therefore, it should be assumed that a mycoplasma-contaminated cell line may be significantly influenced in every respect, and, thus, experimental data derived from such a cell line is likely to be invalid. Contamination is not obvious, either macroscopically or microscopically; thus, routine mycoplasma testing is essential for any cell culture laboratory. Many of the testing procedures developed so far are time-consuming, expensive, inconclusive and unsuitable for screening large numbers of test specimens. This study compares DNA staining, enzyme-linked immunosorbent assay (ELISA), polymerase chain reaction (PCR) and PCR ELISA, to determine which is the best procedure for routine assessment of cell cultures. All four methods gave reproducible results with both infected and non-infected cell lines. Both ELISA methods were easy to perform, reproducible and easily interpreted.