A few hydrophobic grains in otherwise hydrophilic sand render the soil hydrophobic and can completely alter the flow of water through unsaturated sands. In this paper we examine whether percolation theory can explain the phenomenon. Percolation theory has been used to describe the dependence of large-scale flow phenomena on heterogeneities found at the pore scale and should, therefore, be able to explain the water flow behaviour in hydrophobic soil. We show that the theory is valid, in general, for a hydrophilic soil into which a small but increasing fraction of highly hydrophobic grains is mixed. However, the application of percolation theory is limited by the complex interactions of matric potential and contact angle effects due to the introduction of hydrophobic particles.