This work concerns the fatigue crack growth behaviour of nickel base single crystal superalloys. The main industrial application of this class of materials is in gas turbine blades, where the ability to withstand severe mechanical loading in combination with high temperatures is required. In order to ensure the structural integrity of gas turbine blades, knowledge of the fatigue crack growth behaviour under service-like conditions is of utmost importance. The aim of the present work is both to improve the understanding of the crack growth behaviour of single crystal superalloys and also to improve the testing and evaluation methodology for crack propagation under thermomechanical fatigue loading conditions. Single crystal superalloys have anisotropic mechanical properties and are prone to localization of inelastic deformation along the close-packed planes of the crystal lattice. Under some conditions, crystallographic crack growth occurs along these planes and this is a complicating factor throughout the whole chain of crack propagation life simulation; from material data generation to component calculation. Fatigue crack growth testing has been performed, both using conventional isothermal testing methods and also using thermomechanical fatigue crack growth testing. Experimental observations regarding crystallographic crack growth have been made and its dependence on crystal orientation and testing temperature has been investigated. Quantitative crack growth data are however only presented for the case of Mode I crack growth under isothermal as well as thermomechanical fatigue conditions. Microstructural investigations have been undertaken to investigate the deformation mechanisms governing the crack growth behaviour. A compliance based method for the evaluation of crack opening force under thermomechanical fatigue conditions was developed, in order to enable a detailed analysis of the test data. The crack opening force evaluation proved to be of key importance in the understanding of the crack driving force under different testing conditions. vi vii