Previous studies showed that the surface heating patterns of the MA-151 applicator on a 2 cm fat and 10 cm thick muscle phantom had center heating at 581 and 930 MHz and two hot spots near the edges of the applicator at 657 and 779 MHz. The hot spots at 657 MHz were consistent with two blisters on a patient's thigh. Since the heating patterns on muscle only showed good center elliptical heating at all frequencies, in this study we have investigated the effects of fat thickness on the heating patterns. Thermograms of fat and muscle surfaces were taken on phantoms with 0, 0.25, 0.5, 1, and 2 cm thick fat exposed to 631 or 915 MHz energy. The 631 MHz was selected to provide reasonable energy coupling for all phantoms. At 631 MHz, two hot spots were evident on all fat surfaces. The pattern on the muscle surface under the 0.25 cm fat did not show two hot spots, but the heating was elongated in the E-field direction. At 915 MHz, the heating was elongated on the surface of the 0.25 and 2 cm fat, and two hot spots were observed on the 0.5 and 1 cm fat surfaces. However, the muscle heating was elliptical in all cases. The ratio of muscle to fat heating decreased as the fat thickness increased. At 0.5 cm fat the ratio was about 1. These results indicate that fat thickness influences heating in muscle. During treatment with this applicator, surface temperature probes should be placed over potential hot spots. Surface cooling is desirable for heating tumors beneath the fat.