The purpose of this study was to compare misfit-induced strains on implant-supported superstructures fabricated by two impression techniques and two different elastomeric impression materials. A master cast hosting four Straumann implants was constructed. On this cast, a total of 21 implant-level impressions were made by the direct technique using a polyether impression material and synOcta screwed aluminum impression caps (PE-D), and by the indirect technique using polyether (PE-IN) or polyvinylsiloxane impression material (VPS-IN) with snap-on impression caps and synOcta plastic positioning cylinders. Two casts were randomly selected from each group of seven, and a total of four screw-retained superstructures, supported by either two or four implants (one of each type on both casts), were cast in a gold alloy for each group. Linear strain gauges were bonded on the superstructures, and misfit-induced strains were recorded during superstructure connection on each of the working casts and on the master cast using a data acquisition system and corresponding software at a sample rate of 10 kHz. Connection on the implants in the master cast increased strains considerably on most of the superstructures, in comparison with strain gradients measured when the superstructures were connected on the casts from which they were fabricated (P .05). The differences in strain amplitude between connection on the cast from which the superstructure was fabricated and on the master cast were higher for superstructures fabricated by PE-D than for those fabricated by PE-IN and VPS-IN. The snap-on indirect impression technique for Straumann implants leads to acceptable superstructures, regardless of the impression material used.