Statement of problem : There have been previous studies about considerable variations in machining accuracy and consistency in the implant-abutment-screw interfaces. Purpose : The purpose of this study was to evaluate the machining accuracy and consistency of implant/abutment/screw combinations on two randomly selected implants from each of four manufactures. Material and methods : In this study, screws were respectively used to secure a cemented abutment, to a hexlock implant fixture ; teflon coated titanium alloy screw(Torq-Tite) and titanium alloy screw in Steri--Oss system, gold-plated gold-palladium alloy screw(Gold-Tite) and titanium alloy screw in 3i system gild screw ana titanium screw in AVANA Dental Implant system, and titanium screws in Paragon System. The implants were perpendicularly mounted in polymethyl methacrylate autopolymerizing acrylic resin block(Orthodontic resin, Densply International Inc. USA) by use of dental surveyer. Each abutment screw was secured to the implant with recommended torque value using a digital torque controller. Each screw was again tightened after 10 minutes. All samples were cross sectioned with grinder-polisher unit(Omnilap 2000 SBT Inc) after embeded in liquid unsaturated polyester (Epovia, Cray Valley Inc) Results : There were the largest gaps in the neck areas of screws in hexagonal extension implants which were examined in this study. The leading edge of the abutment screw thread (superior surface) was in contact with the implant body thread, and the majority of the contacting surfaces were localized to the middle portion of the mating threads. Considerable variation in the contacting surfaces was noted in the samples evaluated. Amounts of contact in the abutment screw thread were larger for assemblies with Gold-Tite screw, gold alloy screw. Torq-Tite screw than those with titanium screws. The findings of intimate contact between the screw and screw seat were seen in all samples, regardless of manufacturers. However, microgap between the head and lateral neck surface of the screw and the abutment could be dectected in all samples. The findings of intimate contact between the platform of the implant and the bottom of the abutment were consistent in all samples, regardless of manufacturers. However, microgaps between the lateral surface of external hex of the fixture and the abutment could be dectected in all samples. Conclusion : Considerable variations in machining accuracy and consistency were noted in the samples and the implant-abutment-screw interfaces were incomplete. From the results of this study, further development of the system will be required, including improvements in pattern design.