Research indicates that certain yeast strains are beneficial in their capacity to stimulate key microbial populations. This stimulation is strain specific with similar yeast strains exerting their effect on totally different microbial populations. Future yeast culture supplements may contain mixtures of different strains designed to suit specific diets. This, therefore, requires the development of a rapid sensitive technique to differentiate among taxonomically similar yeast strains in animal diets. This technique, termed the Randomly Amplified Polymorphic DNA (RAPD) assay, is based upon the use of randomly designed short polynucleotide primers to amplify genetic sequences from the DNA of the desired yeast strain. Our objective involves the development of this technique to distinguish between closely related yeast strains present in feed. The feed sample investigated was a standard cattle ration containing three strains of Saccharomyces cerevisiae (1026, 2045 and 2020) and Candida utilis 3001 at a concentration of 106 CFU/g respectively. Isolation of single colonies of yeast strains present was achieved by feed extraction in dilution buffer followed by plating a series of dilutions on rose-bengal agar. Thirty randomly selected colonies were cultured in YPD (1% yeast extract, 2% peptone, 2% glucose) broth for 24 - 30 hours at 30°C. Genomic DNA was isolated from yeast cells by standard methods based on subjection of the cells to vortex mixing in the presence of glass beads, triton X-100, sodium dodecyl sulphate, phenol and chloroform. Isolated DNA from randomly selected colonies was amplified by Polymerase Chain Reaction (PCR) for 45 cycles of 1 min at 94°C, 1 min at 36°C and 1 min at 72°C using randomly designed 10 bp primers.