Vegetation monitoring, recording the recovery or change in plant cover over time, for several Karoo shrubs was undertaken to evaluate the Savory Grazing System on a farm in the Succulent Karoo. This evaluation was quantified by establishing if the proposed 120- day rest period between grazing events was sufficiently long enough for complete recovery of the vegetation. Complete recovery was described as sufficient regrowth after a grazing event so that continual cover loss would not result over time. Optimum resting periods would ensure that cover loss, due to grazing, could be recovered so that plant size and reproductive potential was not detrimentally affected. The ellipse intercept method was used to sample piospheres (or zones of attenuating animal impact) around water points in two vegetation types. From these results, 320 line transects for vegetation monitoring were located between 140 and 180m from the water points. One and a half year old piospheres of one farm were compared to seventeen year old piospheres on another farm with similar vegetation composition. The piospheres on the younger farm were expanding at an approximate linear rate of 80m per year, however, this rate slows considerably, as was found on the farm with older piospheres. Stability or equilibrium appears to be reached at approximately 320m from the water point in the older system. The line transects were used to record the change in cover of palatable and unpalatable Karoo shrubs on a monthly basis over an 18 month period. In all instances it was concluded that the proposed 120-day rest period was not sufficiently long enough for complete recovery of the vegetation. The highly palatable species, especially Osteospermum sinuatum and Tetragonia spp., were the most heavily utilized and detrimentally affected. Larger individuals of the palatable shrubs O. sinuatum and Tetragonia spp. were more severely grazed than smaller individuals as a result of smaller individuals utilising spiny nurse plants under which to establish. If the present rest period is continued the result could be overgrazing and local extinction of these important fodder species. The responses of two highly palatable species, O. sinuatum and Tetragonia spp., were monitored closely in relation to rainfall and grazing events. Grazed and ungrazed (protected) individuals were monitored over one year to substantiate the results obtained from the line transects. The rest period was again found to be too short for full recovery and was also found to be reducing the reproductive output of O. sinuatum and Tetragonia spp.. The recruitment of O. sinuatum and Tetragonia spp. was also measured by comparing seedling establishment in grazed, ungrazed (vegetation protected from sheep grazing) and cleared vegetation. In an attempt to rehabilitate these rangelands, Pteronia pallens, a dominant unpalatable shrub, was cleared and the resultant seedling recruitment of O. sinuatum and Tetragonia spp. monitored and compared to grazed and ungrazed vegetation. Juvenile recruitment of these species was found to be significantly lower in the grazed than in the ungrazed vegetation. Recruitment of Tetragonia spp. was found to be significantly higher in the cleared strip compared to the grazed and ungrazed vegetation. The results obtained in this study suggest that the present grazing regime is having a detrimental effect on the vegetation and that revised management procedures are needed to ensure the conservation of these rangelands. The rest periods between grazing events need to be lengthened as well as a reduction in stock numbers. Certain camps need to be skipped on a seasonal basis during the flowering season in order to increase the reproductive output of highly palatable species. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 1995.