SINCE the work of Flörke1 it has generally been accepted that natural opaline silicas fall into two broad categories: (1) specimens which give X-ray patterns indicative of an obviously, although in some cases poorly, crystalline structure; (2) specimens which give X-ray patterns showing only a few diffuse bands. Earlier work by some of us2 supported these conclusions, and a comprehensive re-examination of natural opals has confirmed that the structure can vary from almost perfect α-cristobalite to apparently amorphous or near amorphous material. The X-ray diffractometer traces shown in Fig. 1 illustrate the variations encountered; it should be noted, however, that, of the diffuse bands visible on powder photographs, only that at about 4.1 Å is recorded in traces G and H. Most specimens which would be termed `common opal' by mineralogists are crystalline. Rarely, these may be well crystallized, as in sample B (which may be compared with A, a synthetic α-cristobalite), or very poorly crystallized, as in F. Most samples, however, give traces similar to C, D and E. Most precious opals and their associated `poteh', as well as hyalite, give patterns like G and H.