“…In fact, early discussions of exchange spring behavior in Nd 4 Fe 78 B 18 , such as that of Coehoorn et al [9] or those reported by Kneller and Hawig [7], report the presence of majority amounts of soft magnetic phases a-Fe and Fe 3 B in those samples. However, a recent Rietveld analysis of X-ray and neutron diffraction data for some Pr 4.5 Fe 77 B 18.5 samples [10] showed that Pr 2 Fe 14 B and Fe 3 B were the main phases, with Pr 2 Fe 14 B being dominant (60 wt%). If one were to admit the presence of phases which are non-magnetic at room temperature, such as PrFe 4 B 4 or Pr, then our Mo¨ssbauer data (and the ternary diagram) could be consistent with as little as 10 at% of Pr 2 Fe 14 B in the sample of Fig.…”