This thesis describes a theoretical and experimental investigation of hole injection and conduction in anthracene crystals. Bulk conduction - space charge limited (SCL), from double injection or photogenerated - in this wide (3.7 eV) band gap insulator is dominated by traps in the forbidden gap. A considerable portion of this thesis is devoted to a theoretical study of the effects of various trap distribution on SCL conduction and saturation currents, and to an experimental determination of trapping parameters. It is shown that in principle the trap distribution can be unambiguously deduced from an experimental current voltage curve, and the conditions under which the method fails because of experimental error are discussed. This important theorem has previously only been considered under the step-function approximation to the Fermi-function (Rose, 1955). Calculation of the SCL current-voltage for a Gaussian trap distribution shows that this distribution does not behave like an exponential distribution as proposed by Silinsh (1970). Experimental determinations of trap distributions nearly always reveal an exponential trap distribution. The parameters Hand Tc of this distribution appear to be quite well correlated, although they vary widely from crystal to crystal, irrespective of purity. For very small values of T /T (near unity), H is found to exceed the molecular density and so cannot be interpreted as the total trap density. In some crystals the exponential distribution is seen to be truncated at low energy, in agreement with Sworakowski's (1970) theory. Neither of these observations has been previously reported. A limited theoretical study of SCLC with spatially non-uniform shallow traps is presented to demonstrate the general effects of non-conformity. The predicted dependence of current magnitude on direction of flow is qualitatively confirmed in a photo-SCLC experiment. Around the transition from bulk-controlled SCLC to surface controlled saturation the current is determined by both space charge and the field dependence of the injection mechanism. Study of the latter requires a knowledge of the field at the surface, F0, and a simple method is derived for calculating F0 from an experimental current-voltage curve. The method is independent of the injection mechanism and only requires knowledge of the slope of the log j - log V curve in the SCLC region, or of the trap distribution. The effects of spatially non-uniform traps are shown to be small. Helfrich's (1964) photo detrapping theory is presented in a slightly more rigorous manner, and is extended to show that j ∝ v2results from photodetrapping at an arbitrary trap distribution. A general model of trapping, detrapping and recombination is presented in an attempt to interpret the intensity dependence of an ohmic region of photocurrent-voltage curves which is observed at low fields. This region is thought to result from trap-to-band photogeneration at carrier pairs. The wide variety of intensity dependences predicted by the model renders a detailed interpretation in terms of trap distributions and recombination parameters difficult. The model can be applied to any two-carrier conduction process and is used to investigate the range of behaviour possible in double injection in an insulator with traps. The phenomenological double injection theory developed is confirmed for a simple case by an exact calculation. The remainder of the thesis is concerned with a study of saturation currents, and relating these to hole injection mechanisms from aqueous solutions. The weak interaction theory of charge transfer from electrolytes to insulators, developed by Gerischer (1972) and Mehl and Hale (1967), is briefly reviewed. It is pointed out that this theory has not been verified for organic insulators. Experimental study of the saturation of dark injection from KI3, SnC14and Ce(so4)2 solutions reveals respectively Schottky emission, Poole-Frenkel emission and a combination of Schottky emission and a mechanism which gives results agreeing with the weak interaction theory. Photo injection from water also involves two processes: one behaving like a finite reservoir electrode, as expected from accepted theories (Mulder, 1968) and an unidentified field-assisted emission process. The saturation current is proportional to light intensity and the photovoltage increases as the 0.2 power of light intensity. Both of these saturate at high intensity.