The effects of transportation and delay in processing of blood samples on the concentration of biomarkers are significant in epidemiological studies for which specimens are collected from participants at locations other than a designated center or laboratory. These sources of variability in measurement were studied by collecting two sets of blood samples from 51 men between 26 and 50 years of age. The first set was sent immediately to the laboratory for processing. The second set was transported by car for one hour and then returned to the laboratory for processing. Both sets were stored together at -80 degrees C until the end of the study. Several blood constituents were evaluated. Vitamins, liver enzymes, and electrolytes showed no changes in concentration after transport by car for one hour. There were decreases in the concentrations of red and white blood cells, high-density-lipoprotein cholesterol, glucose, and creatinine after transportation. The transported total cholesterol, total testosterone, free testosterone, alkaline phosphatase, total bilirubin, and thiobarbituric acid-reactive substances increased in concentration. Although transportation and delay in processing of blood samples do not appear to greatly impact relative risk estimates, epidemiologists should be aware of these potential sources of variability in measurement and consider the consequences in their particular study.