The use of veterinary drugs continues to be a challenge since some farmers use drugs for medicinal purposes, while others use them as growth promoters without observing the set withdrawal times. In emerging economies, challenges are experienced in residue testing to ensure residue free animal products. It is therefore necessary to develop a simple, environmentally friendly and cost effective extraction method for veterinary drugs residues. A dispersive liquid-liquid micro-extraction method for simultaneous determination of chloramphenicol, florfenicol and thiamphenicol residues in water, bovine muscle and milk was developed optimised and validated. Analysis was performed with liquid chromatography coupled to electron spray ionization tandem mass spectrometry in negative ion multiple reaction monitoring mode. Dispersive liquid-liquid micro-extraction method development involved optimisation of parameters such as type and volume of extraction solvent, type and volume of disperser solvent and pH. The optimum extraction solvent was dichloromethane at a volume of 250 μL while the optimum disperser solvent was acetonitrile (1 mL) Extraction was found to be optimum at pH 7. The developed method was validated according to Commission decision 2002/657/EC. Linearity. Linearity was observed for water, bovine milk and muscle in the ranges of 0.3 μg kg-1 to 0.9 μg kg-1, 25 μg kg-1 to 100 μg kg-1 and 50 μg kg-1 to 200 μg kg-1 for chloramphenicol, thiamphenicol and florfenicol respectively with regression coefficients ranging between 0.9941 and 0.9999. Limit of detections for CAP, FFC & THAP were 0.082, 3.31, and 2.21 μg kg-1 respectively while Limit of quantifications were 0.163, 7.51 and 6.84 μg kg-1 respectively. The recoveries for water, bovine milk and bovine muscle ranged between 87.2% and 102%. In this study the validated dispersive liquid-liquid micro-extraction has been found to perform very well not only in the water matrix but also in the more complex matrices like bovine milk and muscle. / Chemistry / M. Sc. (Chemistry)