We are reporting the effect of a cellulose acetate 0.20 micron filter (Flow Pore D26) on preparation of platelet poor plasma (PPP) for subsequent assay of platelet specific proteins. PPP was obtained using the Edinburgh anticoagulant mixture and a centrifugation at 2,300 g for 20 minutes. The mean levels of platelet factor 4 (PF4) and beta-thromboglobulin (beta TG) before filtration were: PF4 15.7 ng/ml (1.5-42.8 ng/ml); beta TG 62.0 ng/ml (25-140 ng/ml). After filtration we obtained: PF4 4.8 ng/ml (1.0-8.9 ng/ml), beta TG 28.4 ng/ml (8.5-51.1 ng/ml). Thus difference for both the platelet specific proteins was statistically significant (p less than 0.005). The contamination by platelet-like material was also greatly affected by the filter (before filtration 4.3 x 10(6)/ml), (2-15 x 10(6)/ml), after filtration 2.0 x 10(6)/ml (1-5 x 10(6)/ml), p less than 0.005). If a high speed centrifuge is not available, the use of this simple filter method could avoid falsely elevated values of platelet specific proteins.