The internal structure of the continental lithosphere holds the key to its creation and development, and this internal structure can be determined using appropriate seismic and electromagnetic methods. These two are complementary in that the seismic parameters usually represent bulk properties of the rock, whereas electrical conductivity is primarily a function of the connectivity of a minor constituent of the rock matrix, such as the presence of a conducting mineral phase, e.g. carbon in graphite form, or of a fluid phase, e.g. partial melt or volatiles. In particular, conductivity is especially sensitive to the top of the asthenosphere, generally considered to be a region of interconnected partial melt. Knowledge of the geometry of the lithosphere/asthenosphere boundary is important as this boundary partially controls the geodynamic processes that create, modify, and destroy the lithosphere. Accordingly, collocated seismic and electromagnetic experiments result in superior knowledge than would be obtained from using each on its own. This paper describes the state of knowledge of the continental upper mantle obtained primarily from the natural-source magnetotelluric technique, and outlines how hypotheses and models regarding the development of cratonic lithosphere can be tested using deep-probing electromagnetic surveying. The resolution properties of the method show the difficulties that can be encountered if there is conducting material in the crust. Examples of data and interpretations from various regions around the globe are discussed to demonstrate the correlation of electromagnetic and seismic observations of the lithosphere-asfhenosphere boundary. Also, the observations from laboratory measurements on candidate mineralogies representative of the mantle, such as olivine, are presented.