Creep tests of 117 compact bovine bone specimens were conducted at three temperatures (25, 37, and 43 degrees C), with applied stresses between 71 and 115 MPa. Following testing, the amount of secondary haversian bone in the gage region of the specimens was estimated. The resulting steady-state creep rates (epsilon) were fit to an Arrhenius (e-Qc/RT) model (where Qc is the activation energy for the mechanism(s) controlling creep, R is the gas constant, and T is the absolute temperature) of the type used to describe the classic steady-state creep behavior of metals, ceramics, and metamorphic rocks. The empirical model developed was epsilon = 5.6 x 10(-9) e4.6F sigma 5.2 e-5330/T, where epsilon is the estimated mean steady-state creep rate, F is the volume fraction of secondary haversian bone, sigma is the applied stress, and T is the absolute temperature. There was a positive, significant association between the estimated mean steady-state creep rate and F, sigma, and T. Qc was determined to be 44.3 kJ mol-1, a reasonable value when compared to activation energies for creep in ceramics. It is hypothesized that permanent deformation during creep of compact bovine bone is primarily due to damage mechanisms associated with dislocations in the hydroxyapatite mineral lattice structure.