The study aimed at determining an appropriate sampling design for monitoring the quality of mountain zebra habitat. The parameter used for monitoring was an index of habitat suitability. The value of this index was greater than 20 in the habitat that was most favoured by the mountain zebras, whereas values below 20 were characteristic of moderate to poor habitat. It is recommended that if the index in the most favoured habitat declines to below 20, management intervention in the form of a reduction in stocking rate of large herbivores should be considered. A sample size of 20 randomly located monitoring plots should be adequate to detect a decline to below the critical level. There was considerable patchiness in the distribution of grazing pressure within plant communities, and this needs to be taken into account in the design of the monitoring programme.