The RF B 1 distribution was studied, theoretically and experimentally, in phantoms and in the head of volunteers using a 3 T MRI system equipped with a birdcage coil. Agreement between numerical simulation and experiment demonstrates that B 1 distortion at high field can be explained with 3D full-Maxwell calculations. It was found that the B 1 distribution in the transverse plane is strongly dependent on the dielectric properties of the sample. We show that this is a consequence of RF penetration effects combined with RF standing wave effects. In contrast, along the birdcage coil z-axis the B 1 distribution is determined mainly by the coil geometry. In the transverse plane, the region of B 1 uniformity (within 10% of the maximum) was 15 cm with oil, 6 cm with distilled water, 11 cm with saline, and 10 cm in the head. Along z the B 1 uniformity was 9 cm with phantoms and 7 cm in the head. The recent development of high-efficiency head gradient coils and ultrafast MRI pulse sequences has allowed an impressive number of functional MRI studies in human brain research to be realized. The need for higher sensitivity has pushed current MRI research towards the development of high field (Ͼ1.5 T) MRI systems (1). However, magnetic field inhomogeneity and susceptibility artifacts increase with field strength, requiring the development of methods to evaluate and, possibly, eliminate their effects (2). Even if the problems of B 0 magnetic field susceptibility were to be solved, the effect of RF magnetic field (B 1 ) inhomogeneity has long been recognized as a potential source of image artifacts in high field systems (3,4). In fact, for frequencies higher than 64 MHz the RF eddy currents induced in the human body cannot be neglected. Moreover, because the effective wavelength of the RF field is comparable to or smaller than the dimension of the human body, there is a significant variation of the B 1 phase along the sample (3). This gives B 1 standing wave effects and consequently the RF field homogeneity can be strongly degraded. As a further consideration, at high field the distribution of the power deposition caused by the RF irradiating field must be carefully considered for safety reasons and strict limits on the specific absorption rate (SAR) have been established (5,6).Because of the complex anatomical structure of the human head, it is impossible to analytically evaluate the RF B 1 spatial distribution. In the past few years, for RF coils operating between 64 MHz and 341 MHz, numerical electromagnetic (EM) computational techniques have been developed to evaluate the B 1 and the SAR in phantoms and in the human head (7-13). Although very powerful, these numerical techniques use a simplified model of the human head (typically, 2 mm voxel, up to 20 tissues, measured or interpolated/extrapolated dielectric properties of tissues). The effects of head size, position, and motion within the RF coil have not been considered. Additionally, they make many assumptions about the RF coil modeling (intrinsic RF B 1 distribution, coi...