Harvesting of sandalwood (Santalum spicatum) occurs mainly from wild stands in the semi-arid and arid regions (typical annual rainfall 150–300 mm) of Western Australia. Regeneration of wild sandalwood in these regions is believed to be low since the occurrence of changes in land use associated with European settlement. This is thought to be due to factors including drought, poor seed dispersal and grazing. The objective of the study was to increase the germination and establishment of sandalwood through exploring seed response to rainfall. Additionally, the potential of soil-preparation techniques to utilise trace amounts of moisture was investigated. Two 25-ha plots were located either side of the semi-arid–arid divide. Into the plots 16 replicates of the control and 96 replicates of treatments were installed and sown with 11 200 seeds. Rainfall and other weather parameters were recorded at each site with an automated weather station. The study was replicated in 2008 and 2009. It was concluded from the study that there was a statistically significant relationship between germination and rainfall. It is proposed that the germination threshold is 264 mm per year which coincides with the long-term annual rainfall average of Kalgoorlie, Western Australia. Furthermore, a statistically significant relationship between germination and soil preparation was demonstrated. Ripping crust-forming soils before sowing and the construction of water-harvest banks had a positive effect. Information gained from these studies has led to the Western Australian State Government implementing a seeding program to increase sandalwood regeneration in the semi-arid region.