“…The chemical variability of natural amphiboles gives rise to many overlapping absorption bands in the principal OH-stretching region of the infrared spectrum, and the spectra commonly do not contain enough information to allow the spectra to be fit correctly. This problem may be overcome by using (1) synthetic amphiboles (Raudsepp et al ., 1987 a , b , 1991; Boschmann et al ., 1994; Della Ventura et al ., 1996 a , b , 1997, 1998 a , b , 1999, 2001, 2003; Hawthorne et al ., 1997, 2000; Robert et al ., 1999, 2000; Najorka and Gottschalk, 2003), the compositions of which are usually sufficiently constrained that a combination of Rietveld refinement or single-crystal refinement, infrared and MAS NMR spectroscopy, and local bond-valence arguments are sufficient to derive short-range order arrangements, or by using (2) gem amphiboles (Tait et al ., 2001; Abdu and Hawthorne, 2009; Heavysege et al ., 2015) for which the compositions are commonly simple. Thus the majority of work on SR-OD in amphiboles involves interpretation of the spectra of synthetic and gem-quality samples where the composition either can be controlled or is simple enough to result in spectra that can be interpreted rigorously.…”