||Reinforced concrete shear walls are frequently used in multi-storey buildings exposed to high lateral loading originating from wind or seismic disturbances because of their high in-plane stiffness. The shear walls are usually pierced by a number of openings dividing them into individual solid walls, which are normally interconnected by strong coupling beams. The beams usually span over short distances and are relatively deep. In the seismic design of tall buildings, in order for a coupled shear wall structure to develop full strength and exhibit ductile behaviour, it is of primary importance that coupling beams possess the considerable ability to deform plastically. An accurate prediction of both ultimate strength and ductility capacity is required. The work in this thesis covers an experimental programme and the theoretical development for shear capacity of deep steel fibre reinforced concrete (SFRC) coupling beams. Experimental study is first conducted to study the behaviour and shear strength (both ultimate and post-cracking) of the deep SFRC coupling beams under both monotonic and reversed-cyclic loads, respectively. A new experimental rig to test coupling beams been proposed, where a very unique feature is that the loading is applied with simple force paths to the beams through the rotation of the walls. The span-to-depth ratio of specimens can be adjusted in a wide range by changing the span rather than the depth of the beam. This allows a direct comparison of results of different specimens without considering the size effect. The loading frame restraints the horizontal deformation of the beam, thus modelling the behaviour of a real coupling beam restrained by the floor slabs in buildings. In the test programme, fourteen large-scale coupling beam specimens are constructed, where seven are tested under monotonic loading while the other are tested under reversed cyclic loading. The important variables in the tests include the concrete matrix used (i.e. steel fibre concrete or pure concrete), span-to-depth ratios of the beams and ratios of transverse steel reinforcement. All fourteen specimens tested failed in a brittle shear manner; however the strength drop is less severe in the case of the steel fibre reinforced specimens. The test results show the significant improvement in the post-peak behaviour of the coupling beams after the addition of steel fibres (by as much as 50%). It has also been shown that the ultimate shear strengths of the specimens tested are much higher than those predicted by the current codes of practice including BS 8110 and Eurocode 2. An analytical model has been developed. The model uses the principles of solid mechanics, fracture mechanics as well as the 1st and 2nd laws of thermodynamics (i.e. energy methods). The proposed model is capable of accurately predicting the ultimate shear capacity and load-displacement behaviour of reinforced concrete and fibre reinforced concrete deep beams. The steel fibre contribution is considered as a concrete property (i.e. tensile strength) and is assumed plastic. The model of concrete softening is based on the model empirically developed and implemented in the Modified Compression Field Theory. A series of analyses has been carried out for comparing the results of the proposed model with those obtained from a wide range of experimental work on reinforced concrete and fibre reinforced concrete deep beams. The results show a very good agreement. It has been shown that the proposed model is capable of accurately predicting the shear capacity and load-displacement behaviour of both reinforced concrete and fibre reinforced concrete deep beams.