Soil Moisture and Soil Electrical Conductivity (EC) are important parameters for data-driven farming. This knowledge can help a farmer improve crop yield, reduce input costs, and adopt sustainable agriculture practices. However, the high cost of commercial soil moisture and EC sensors has limited their adoption. In this paper, we present the design and implementation of a system, called SMURF, that senses soil moisture and soil EC using RF propagation in existing Wi-Fi bands. It overcomes the key challenge of limited bandwidth availability in the 2.4 GHz unlicensed spectrum using a novel multi-antenna technique that maps the propagation time and amplitude of Wi-Fi at the different antennas as a function of the refractivity and permittivity of soil, and uses them to infer soil moisture and EC. Our experiments with software defined radios (USRP and WARP), and two commodity Wi-Fi cards show that SMURF can accurately estimate soil moisture and EC using Wi-Fi, thereby enabling a future in which a farmer with a smartphone that has a Wi-Fi radio can sense soil in her farm without investing 100s of dollars in soil sensing equipment.