Surrounding this fine martensite there are fine austenite grains that are able to grow since austenite does not
consume martensite. Probably that is the reason why the reverted austenite grains at 700 °C are larger than those ones observed at lower temperatures. From Fig. 9(c) it is also noticeable the presence of larger recrystallized austenite grains. These grains are consequence of recrystallization of the austenite that did not experience martensitic transformation during deformation. The combination of discontinuous recrystallization and growth of reverted austenite explains the highest decrease of hardness between 600 and 700 °C depicted in Fig. 8(b). Finally, Fig. 9(d) shows that the microstructure of the sample annealed up to 800 °C (piece 4) is practically austenitic and the fraction of α0-martensite is
negligible. It also must be noticed that the tiny red regions in this figure can be δ-ferrite. It corroborates the low value of Ms found for this sample (see Fig. 8(a)). Besides, the hardness value of this sample (see Fig. 8(b)) is close to the correspondent value of the asreceived material.