Magnetic resonance imaging (MRI) has shown the potential to provide information related to the microarchitecture of the trabecular bone matrix. Detailed measurements of the structure of bone samples were made by histomorphometry and compared to T2' (%CV=6.3 ± 2.7 - 8.7 ± 5.0) measured with a specially implemented sequence, PRIME (Partially Refocused Interleaved Multiple Echo), at 1.5 Tesla. After correction for bone mass, T2' in 12 samples was significantly correlated to trabecular perimeter (r=-0.61,p=0.04), number density (r=-0.61,p=0.03) and width (r=-0.66,p=0.03), demonstrating the ability of MRI to characterise these important measures of the trabecular bone surface. This result strengthens the hypothesis that the sensitivity of T2' to osteoporosis-related bone changes is due to magnetic susceptibility effects in which rapid transitions between bone and marrow create local magnetic field inhomogeneities which result in a shortening of the T2' relaxation time. In-vivo measurements of bone mineral density (BMD) and ultrasound parameters were made in the calcaneus of 55 postmenopausal women aged 43-87 years. Calcaneal T2' was measured in 32 women and was significantly correlated to BMD (r=-0.79,p<0.0001) and ultrasound parameters (r=-0.59,p=0.0004). These 3 techniques have not previously been compared in the same study population. Z scores for all techniques were significantly different in normal and osteopenic women defined by spine BMD. All BMD and ultrasound parameters were significantly different in normal and osteopenic women after adjustment for age and years since menopause (YSM), while T2' was only significantly different before correction. T2' was significantly correlated to the number of vertebral fractures (r=0.38, p=0.03), as were BMD and ultrasound. The only parameters that were significantly different (after adjustment for age and YSM) between women with (n=19) or without (n=14) vertebral fractures were the change in BMD of the calcaneus or lumbar spine over one year. The calcaneus was shown to be a sensitive and precise site for the measurement of BMD (%CV=0.7 ± 0.4 - 1.3 ± 0.6).