An understanding of the process by which polymerising free radicals are created is of fundamental importance for any type of free radical polymerisation system. In the case of emulsion polymerisation this process is complicated by the fact that free radicals are usually generated in the aqueous phase, while the principal site of polymerisation is the interior of hydrophobic latex particles. This thesis details a study of the kinetics and mechanism for radical entry into latex particles in-emulsion polymerisation. A systematic investigation is carried out identifying the influences of initiator, monomer and particle surface type on the entry efficiency. Experimental entry efficiencies are obtained from a range of seeded styrene emulsion polymerisation systems at 50°C, employing potassium peroxodisulfate and 2,2' -azobis-(2-methylpropionamidine) dihydrochloride as initiators, and using latex particles stabilised by positive or negative surface charge. Changing either initiator or particle type is seen to affect the entry efficiency. Experimental results are able to be rationalised in terms of an existing theory for entry which suggests that entry occurs for persulfate-derived radicals after addition to two styrene monomer units, and for amidinium-derived radicals after addition to only one styrene monomer unit. Entry efficiencies are also obtained from seeded experiments at 50°C using persulfate as initiator and methyl methacrylate as monomer, with considerably greater uncertainty arising in this case due to unexplained variations in the "acceleration" during Interval II of polymerisation. Comparison with entry theory for this system indicates that entry occurs for a persulfate-derived radical only after addition to at least 20 methyl methacrylate monomer units. Attempts to confirm this inferred value by analysis of aqueous-phase oligomers using mass spectrometry and gel-permeation chromatography are inconclusive. In the hope of improving the accuracy with which entry rate, data may be obtained from kinetic experiments the following investigations into more other aspects of emulsion polymerisation kinetics are carried out. The contribution to entry from spontaneous polymerisation is measured for a wide range of styrene and methyl methacrylate systems and various mechanistic inferences are made. The particle surface and aqueous phase are identified as likely loci for spontaneous radical generation. The nature of chain stopping reactions in emulsion polymerisations of methyl methacrylate at 50°C are investigated through kinetic experiments and analysis of polymer molecular weight distributions. The radical chain length distribution is found to be approximately ''transfercontrolled", and both kinetic data and molecular weight distributions are well described by current models for termination and the effects of column broadening in gel-permeation chromatography. A new kinetic model for emulsion polymerisation is developed which extends the Smith-Ewart treatment to take account of all reactions of monomeric and aqueous-phase radicals. This new model treats exit and re-entry in terms of elementary physical and chemical processes, employing microscopic rate parameters and obviating the use of less meaningful parameters, such as k and a, which have traditionally been required for non-zero-one systems. The model is applied to the styrene and methyl methacrylate systems used in experimental work and confirms the accuracy of the entry data obtained using existing methods. However, no improvement in the fitting of methyl methacrylate acceleration data is obtained from using the new model with best estimates for all parameter values. Finally, new approaches to modelling entry are proposed which may overcome the limitations posed by existing entry models. Most importantly, recommendations are made for including the kinetics of adsorption, desorption and entry of all aqueous radicals.