A nested case-control study was undertaken involving men participating in the Multiple Risk Factor Intervention Trial (MRFIT). Serum samples from 712 men, stored for upto 20 years, were analysed for homocyst(e)ine. Cases involved non-fatal myocardial infractions, identified through the active phase of the study, which ended on February 28, 1982, and deaths due to coronary heart disease, monitored through 1990. The non-fatal myocardial infarction occurred within 7 years of sample collection, whereas the majority of coronary heart disease deaths occurred more than 11 years after sample collection. Mean homocyst(e)ine concentrations were in the expected range and did not differ significantly between case patients and control subjects: myocardial infarction cases, 12.6 micromol/L; myocardial infarction controls, 13.1 micromol/L; coronary heart disease death cases, 12.8 micromol/L; and coronary heart disease controls, 12.7 micromol/L. Odds ratios versus quartile 1 for coronary heart disease deaths and myocardial infarctions combined were as follows: quartile 2, 1.03; quartile 3, 0.84; and quartile 4, 0.92. Thus, in this prospective study, no association of homocyst(e)ine concentration with heart disease was detected. Homocyst(e)ine levels were weakly associated with the acute-phase (C-reactive) protein. These results are discussed with respect to the suggestion that homocyst(e)ine is an independent risk factor for heart disease.