A study was undertaken to evaluate the parameters involved in the performance of levitation coils with particular emphasis on the heating effect. The field strength along the axes of various coils was measured using a small search g proportionial to the field strength. The search coil was calibrated with a straight solenoid coil in which the field strength was calculated. By considering each turn of a levitation coil as a short solenoid whose length was the tubing diameter, the field strength contribution from each turn was calculated. These values were summed along the axis to calculate the coil field. To aid these calculations, tables of field strength as a function of axial distance were prepared for various raddii turns to different sizes of tubing. The rate of increase of field strength along the coil axis per ampere of coil current was defined as the coil gradient. A parametric equation was derived relating the power input to a levitated sample to the coil gradient, sample size and resistivity, and frequency. The validity of this equation capacity and time to reach temperature. The contribution from a given turn to the total coil gradient was calculated as a function of its location. Thesemore » data were graphed with axial distance as abscissa and turn radius as ordinate showing positions which give the same values of gradient contribution. The positions which contribute most to the coil gradient could be used to design coils of high gradient and, consequently, good heating efficiency. Other results of the study include a possible method for measuring the surface tension of molten metals and a method for choosing the minimum frequency for efficient heating of a levitated sample. (auth) « less