The objectives of this dissertation are to evaluate and compare a new capacitive sensor, a watt meter and a strain gaged torque sensor for wet granulation monitoring in a high shear mixer. Toward this end, a 10 L vertical high shear mixer was instrumented to monitor power consumption, torque, moisture distribution and capacitive changes during agglomeration. The percent moisture content related linearly to the amplitude channel response. Yield values provided a rheological property to relate with power consumption and torque measurement. In a comparative study of the two most commonly used monitoring techniques, torque measurement was shown to be more sensitive than power consumption measurement, but this increased sensitivity did not offer any advantage over power consumption measurement. In another comparative study the amplitude channel of the capacitive sensor appears to more clearly differentiate between binder levels in hydrous lactose-hydroxypropyl methylcellulose granulations than either power consumption or torque measurement, based on particle size distributions. Four formulations were used to evaluate the effect of varying agitation rate and rate of fluid addition on the granulation endpoint determined by the capacitive sensor. For hydrous lactose (Lactose) with 4% polyvinylpyrrolidone K 29/32 (PVP), dicalcium phosphate anhydrous milled (DCP) with 5% PVP and a one to one mixture of Lactose and microcrystalline cellulose (MCC 102) with 5% PVP, the amplitude channel of the capacitive sensor predicted similar endpoints based on particle size, flow rate and crushing strength-compression force profiles under varying agitation and addition rates. The apparent independence of the amplitude channel to variations in agitation rate and rate of liquid addition suggests that this component of the capacitive sensor may be scaled up. The amplitude channel was successfully scaled up from a 50 L to a 130 L horizontal high shear mixer based on particle size, flow and crushing strength-compression force profiles for Lactose/PVP formulations. In addition, the amplitude channel was able to compensate for differences in moisture distribution when the chopper was employed in the scale-up study. A centrifugation technique, thermogravimetric analysis and nuclear magnetic resonance (NMR) inversion recovery technique were developed to compare the interaction of moisture with each of the filler systems used in the endpoint detection study. (Abstract shortened by UMI.)