A brief history is given of the development of resistivity detection in archaeology, followed by an outline of basic principles and theory, and of the evolution of the Martin-Clark resistivity meters, including circuit diagrams. The chief outstanding problems are identified as choice of the most suitable probe configurations, climatic water balance effects, and speed of working. For studying the configuration problem, the design of a rapidly operating automatic electrolytic tank system is detailed, together with a set of simulated archaeological features. Ten configurations are examined, of which the Twin-Electrode, Double Dipole and Square Array are found most effective, with the addition of Wenner for selected purposes. The Triple Electrode also appears promising. For the climate study, fixed tiaterse lines were established across three silted up ditches of known profile and widely varied size on Upper Chalk in southern England, and measured monthly for 19 months. The pattern of resistivity response with water balance differed greatly according to ditch size. The data, and results obtained by others for Triassic sandstone, London Clay and limestone are collated, and guidelines for interpretation and the choice of optimum survey times, probe configurations and spacings established. Some analysis of the related phenomenon of crop mark production is appended. In combination, the tank and climate studies show that the resistivity anomalies caused by many archaeological features approximate to a horizontal high resistivity lamina. Examples of surveys using the optimum configurations are given, the results being compared with those of magnetic surveys where possible; methods of plotting and processing the data are also compared. Some high performance survey systems incorporating results from this research are described.