Thesis submitted in fulfilment of the requirements for the degree Master of Technology: Horticulture in the Faculty of Applied Sciences at the Cape Peninsula University of Technology, 2013 / Cape Flats Sand Fynbos (CFSF) is one of the most poorly conserved vegetation types in the Cape Floral Kingdom, and a large proportion of unconserved land is degraded, primarily as a result of transformation by agricultural, urban developments and invasion by alien plants. Fynbos restoration is one of the most important management interventions, both within the current conservation areas and in any future land remnants acquired for conservation. Many extant remnants are fragmented and isolated, and if successful restoration protocols are found, it may be possible to improve the conservation targets for this critically endangered vegetation type. On old fields, where indigenous soil seed banks have been lost due to alien plant invasion and anthropogenic action, it is essential to reintroduce the longer-lived fynbos components that contribute to vegetation structure, in order to facilitate the progress of the ecosystem on a more natural trajectory. This research is built on an earlier study of optimal ground-preparation treatments for restoring Sand Fynbos to old fields. This former study indicated that fossorial mammals (molerats and gerbils) may occur in dense colonies on old fields and present an obstacle to successful seedling establishment. The project aims to provide protocols for the establishment of indigenous seedlings from harvested seed onto old fields, in order to restore Sand Fynbos vegetation. Different pre-sowing treatments and sowing techniques were tested on large field plots to determine the most efficient protocol. The objectives of the research were: a) to investigate optimal pre-sowing treatments of indigenous seed for restoring degraded Sand Fynbos vegetation in old field sites; b) to investigate optimal sowing techniques on large field plots for restoring degraded Sand Fynbos vegetation in old field sites with depleted indigenous seed banks; c) to provide guidelines and disseminate information on optimal sowing protocols, and their costs, for restoring Sand Fynbos vegetation in degraded areas and old fields. A further component of the research was to calculate the costs of all treatments on a per hectare basis in order to assess the cost-effectiveness of the different options. Several different seed treatments may potentially increase the germination rate and promote fynbos restoration. These are scarifying, smoke, smoke water, chemical, light and temperature pre-sowing treatments. In order to keep the number of treatments (including their interactions) to a manageable level, only soaking in smoke water extract and seed coat scarification with course sand and grit were tested. A seed sample of each species was x-rayed at the Millennium Seed Bank in the United Kingdom, to test for viability in the seed samples, 52 % of the seed collected were empty, a typical indication of wild harvested seed. All species were germinated at 10/25 °C and 16/8 hours light/dark respectively. Scarification had a larger overall germination success, smoke water had very little effect on CFSF species, it is rather that germination is related to temperatures during a fire that result in seed coat splitting. It was recommended that further investigation using more species across the Sand Fynbos vegetation be conducted on pre-germination effects of heat and scarification. The study site had been cleared of woody invasive alien vegetation and additional site preparation included the application of a systemic herbicide to kill undesirable herbaceous weeds, prescribed fire to clear the site of woody debris and destroy weed seeds, and the local control of fossorial mammals (gerbils and molerats) by placing raptor perches and owl nesting boxes around the site. This research found that the use of herbicide shortly after the prescribed fire and once again prior to sowing was successful in controlling herbaceous weeds and the indigenous grass Cynodon dactylon. The challenge to using prescribed burning on old fields was low fuel loads, which resulted in a cool patchy. It is suggested that cutting and spreading of alien plant biomass is tested as a solution, however, the material must be evenly spread across the site and not stacked into piles which can cause excessively hot fires and scorching of the soil. In order to better understand soil conditions across the site, soil samples were collected prior to sowing, to analyse for soil macronutrients, organic matter and pH. It was established that all the excess nutrients added to the site from agriculture and pasturage over the years had leached from the soil. However, the organic matter content of the soil was extremely low and research needs to be done on the organic carbon content of the soils, how these relate to soil micro biota (which species are present and their relationship with CFSF species) and how best to enrich the site with humic matter for restoration and establishment of Cape Flats Sand Fynbos. The field trial was set up in the Blaauwberg Nature Reserve, a random split-plot block design, was replicated five times and used to investigate the selected seedbed preparation and sowing techniques, namely: broadcast sowing onto unprepared seedbed, broadcast with plank embedding of seed onto disked seedbed, broadcast onto disked seedbed and hydro-seed with disked seedbed. Results from the research found that the most successful methods for sowing seeds were the hydro-seeding and broadcast with plank embed. These methods may have provided better contact between the soil and seeds and better protection from predation and wind. Economically the broadcast and embed was better as machinery was more efficient and effective than manual labour. This study recommended that these two methods be combined with the additional planting of rare and threatened species in clumps to determine the benefits and interactions of each technique over the long term.