The focus of this thesis is a fundamental investigation of the process of mechanisms involved in the process of screen printing. Accordingly, the process is sub-divided into three distinct stages: the flooding of excess ink onto the mesh; the metering of the ink; the printing of the image. This differs from the normal one of viewing it as a two- stage process where metering and printing are not distinguished from each other. Experiments are conducted using a purpose built contact screen printing apparatus that allows a detailed study of the effect of each of the press variables on the final print thickness. The results show that the process is relatively insensitive to many of these factors, with the geometry of the mesh (dictated by thread count and diameter) and, to a lesser extent, the fluid properties, being the overriding parameters that affect the print thickness. The work also highlights, for the first time, that the mesh is only approximately half filled with ink prior to printing, and the semi-empirical models used by industry to predict print thickness have been refined and improved accordingly. Computational fluid dynamics (CFD) simulations of the problem have also been performed to highlight the mechanism of ink transfer at the printing stage of the process. The results reveal a free-surface dominated process, one that shows little dependence on many operating parameters, and in accordance with the above complementary experimental investigations. Finally, the process by which fluid fills the screen under the action of a moving blade is studied. An idealised novel geometry is adopted, that of a single cell cut into a smooth plate. A precision test rig and optical viewing system were designed and built for this purpose, with high speed photography used to reveal a rich dynamic process where a number of final states are possible. These include cells that are completely full, partially full or empty of fluid, and the dependence of filling type on the properties of the system has been determined.