Walker, Skinner, Addenbrooke, Rayner, Orlich, Schultze, Thielers and others* have given a great amount of useful information on electrometers for the measurement of dielectric loss. We have found instruments made somewhat after the design of Skinner and Rayner, so rugged, and so useful as laboratory instruments, even when the instrument is subjected to severe jars and mechanical vibration, and capable of such great accuracy that we have spent considerable time developing methods of use and investigating the sources of error. Our instruments are essentially high-voltage instruments as the needle may have impressed upon it voltages up to about 8500 volts. One of these instruments has been in use almost continuously since 1913, a period of ten years, and for the last seven years it has been continually in use without needing repairs of any sort, in spite of breakdowns of the load being measured, and no adjustments have been made except an occasional turning of the quadrant leveling screws. We have developed what we believe to be a new zero method of measuring power factor which has many advantages, and have also used the electrometer as a detector in a high-voltage bridge. We have also gone thoroughly into the errors of the electrometer, both for the deflection and zero methods, and have included these errors in our equations, so that we believe they may be taken care of practically.