Several factors influence the behaviour of infilled frames, which has been studied in the last fifty years. One might assume that so many years of research in one theme might be enough to nearly fully understand it but this is not the case. The new generation of design standards, namely the Eurocode 8 (EC8), impose the use of reinforcement in these walls in order to prevent a brittle collapse and makes the structural engineer accountable for this requirement, yet it fails to provide enough information for the design. Motivated by the stated reasons, this thesis aims at understanding the seismic behaviour of infill walls when designed following the prescriptions of the EC8, therefore reinforced with bed joint and plaster reinforcement, and compares them to the seismic behaviour of the infill walls considered as a standard in the last three decades in Portugal, which is an unreinforced double leaf. For this purpose, three RC concrete buildings were constructed at a scale of 1:1.5 and tested on the shaking table of the National Laboratory for Civil Engineering, Portugal, each with a different infill solution but with the same geometry. Next, the experimental results were compared, in terms of demand and capacity, with the analytical solutions proposed by different authors and design standards. A numerical study was carried out, which includes model updating and non-linear static analysis. Finally, two novel design proposals are presented: one regarding the analytical computation of the natural frequency of infill walls; another regarding the reduction of the out-of-plane capacity of the infill wall due to the presence of openings. The experimental tests show that the double leaf unreinforced infill walls underperformed during a large earthquake, collapsing out-of-plane by rotating as a rigid body around the base line of the model. The infill walls with bed joint and plaster reinforcement did not collapse out-of-plane due their connection to the RC frame. The comparison done between the experimental results and analytical solutions of the design standards show that the formulation proposed by the EC8 presents acceptable results for most of the infill walls in the present work, as far as the out-of-plane demand is concerned. As for the capacity, the EC6 and FEMA 306 provide acceptable results. Both design proposals present a good correlation with the experimental results.