The fatigue crack propagation behavior of Ti–5Al–2.5Fe with various microstructures for biomedical applications was investigated in air and in a simulated body environment, Ringer's solution, in comparison with that of Ti–6Al–4V ELI and that of SUS 316L stainless steel. The crack propagation rate, d a/d N, of Ti–5Al–2.5Fe in the case of each microstructure is greater than that of the Widmanstätten α structure in Ti–6Al–4V ELI in air whereas d a/d N of Ti–5Al–2.5Fe is nearly equal to that of the equiaxed α structure in Ti–6Al–4V ELI in air when d a/d N is plotted versus the nominal cyclic stress intensity factor range, Δ K. d a/d N of the equiaxed α structure and that of the Widmanstätten α structure in Ti–5Al–2.5Fe are nearly the same in air when d a/d N is plotted versus Δ K. d a/d N of Ti–5Al–2.5Fe is nearly equal to that of SUS 316L stainless steel in the Paris Law region, whereas d a/d N of Ti–5Al–2.5Fe is greater than that of SUS 316L stainless steel in the threshold region in air, when d a/d N is plotted versus Δ K. d a/d N of Ti–5Al–2.5Fe or Ti–6Al–4V ELI is nearly the same in air and in Ringer's solution when d a/d N is plotted versus the effective cyclic stress intensity factor range, Δ K eff, whereas d a/d N of Ti–5Al–2.5Fe or Ti–6Al–4V ELI is greater in Ringer's solution than in air when d a/d N is plotted versus Δ K.