Determining the impact of inbreeding on white clover growth will assist in determining the optimal breeding strategies for future population improvement. A dominant self-fertility allele (Sf) at the S locus was used to inbreed white clover to nearhomozygosity (F=0.99). Inbreeding depression was higher in the glasshouse experiment than was observed under sheep grazing in the field. It was also higher for comparable generations with a 54% reduction in herbage yield of the S1 generation in the glasshouse compared with 29% yield reduction under grazing. The level of inbreeding depression for herbage yield of the S1 and S2 generations observed in the glasshouse were consistent with the theoretical reduction in yield. However in later generations the actual herbage yield was consistently higher than expected. This may be due to heterozygosity being maintained during inbreeding or to the elimination of weaker individuals under higher competition and grazing experienced under field conditions. The degree to which inbreeding depression is overcome by outcrossing (heterosis) in white clover will determine whether hybrids or semi-hybrid cultivars can be successfully developed