Three recent studies discussed the possibility that the National Committee for Clinical Laboratory Standards (NCCLS) recommendations that the coagulation specimen should be the second or third tube collected are unnecessary. However, only one reagent/instrument was used in each study. Our protocol differed from the previous studies because we performed the assays on three different reagent/instrument systems on the same samples. Our study used photo-optic, mechanical, and nephelometric systems of clot detection. After obtaining informed consent, we obtained two blue-stoppered tubes of blood from 95 subjects: 15 normal patients and 80 patients currently on coumadin therapy. No discard tube was drawn for coagulation testing. A prothrombin time with an international normalized ratio and an activated partial thromboplastin time, were performed on each tube. Laboratory One used a MLA 1600C (Hemoliance) with Thromboplastin DS (Pacific-Hemostasis, ISI of 1.11) and APTT-LS (Pacific-Hemostasis). Laboratory Two used an STA (Diagnostica-Stago) with Neoplastine CI+ (Diagnostica-Stago, ISI of 1.14) and PTT-LT (Diagnostica-Stago). Laboratory Three used an ACL 300 with Plastinex (Biodata, ISI of 1.67) and Actin FSL (Dade Behring). No clinical or statistically significant differences were seen between the first or second tubes on any of the three reagent/instrument combinations in the PT in seconds, international normalized ratio reporting, or APTT results. Our results indicate that the NCCLS guidelines for obtaining a second tube when performing coagulation testing should be considered for elimination when new revisions are published.