A number of natural kaolinites from a variety of world sources have been investigated using electron spin resonance at both X-band and Q-band. The results show systematic differences some of which are related to the crystallinity of the material, some to the presence of transition metal ions other than Fe3+ and hole-trapping defects.

The results provide evidence for at least three sites for substitution of Fe3+ having large crystal fields : (i) a site with near maximum rhombic character having crystal field parameters D > 1·2, λ = ⅓ the occurrence of which is correlated with lack of crystalline perfection due to stacking defects or to the proximity of surfaces, (ii) a site with less rhombic character with λ = 0·234, D = 0·585 as likely parameters, and (iii) a site with parameters near the values λ = 0·207, D = 0·322.

Mössbauer and ESR evidence suggests that the principal sites of Fe3+ substitution are octahedral. The presence of Fe3+ in adjacent cation sites leads to a very broad resonance centred near geff = 2. Other resonances in the spectra are attributed to the effect of trapped holes, some at least of which are situated at oxygen sites adjacent to Al3+ ions substituting in sites normally occupied by Si4+.

Hyperfine effects due to the presence of the transition ions Mn2+ and (VO)2+ are also observed.

A method is given for calculation of X-ray scattering from interstratified materials which is applicable to components that do not possess centres of symmetry. This procedure allows contributions from interlamellar material to be included in the calculation without any assumptions being imposed by the method on the composition of the components. The method is in fact a simplification of the general formula given by Kakinoki & Komura (1965).

Initial exchange of Cu(II)-montmorillonite by tetrapropylammonium (TPA) or tetramethylammonium (TMA) ions results in segregation of different ions in randomly interstratified layers, with particular preference for TPA over Cu(II) during exchange. More complete exchange reverses this preference and both Cu(II)-TPA and Cu(II)-TMA clays become more homogeneous. The copper ions of the homogeneous Cu(II)-TPA clays are not fully hydrated while the TMA ions have less effect than TPA on the hydration of copper, probably a result of the smaller radius and higher hydration energy of TMA. Methods of X-ray diffraction, thermogravimetric analysis, BET surface area determinations, as well as infrared and electron spin resonance spectroscopy, are used in supplying evidence of cation distribution and hydration properties of these clays. Definite changes in the properties of hydrated copper when associated with tetraalkylammonium ions on the clay surface are indicated.

Investigation of the clay mineralogy of forty-seven samples of sediments from boreholes in the western Nile Delta, an area little studied hitherto, and from surface sites on the mouth of the Nile and adjacent coast shows that the clay fractions consist of dominant iron-rich, dioctahedral, randomly interstratified smectite-illitcs together with kaolinite, illite and chlorite.

Amounts of the constituent minerals of the clay fractions are estimated from their X-ray diffraction intensities, supported by selective dissolution chemical data, and a new method is used to estimate the proportion of expanding layers in randomly interstratified smectite-illite. The results, which confirm and extend the work of previous investigators, also show that there is little correlation between the clay mineral composition and texture of the sediments, only kaolinite being weakly linearly correlated with clay content. Transformation of 2:1 layer silicate minerals occurs within the buried sediments ; chlorite is transformed and smectite and illite interlayers redistributed within randomly interstratified smectite-illites.

Lower Palaeozoic mudstones from Cwmcadian, Gwynedd are dominantly composed of quartz, illite and chlorite, with subordinate albite, orthoclase and rutile. Chemical formulae are determined for chlorite and illite using a combination of chemical and X-ray data. The formulae indicate that the illite is dioctahedral and suggest that the chlorite structure contains one dioctahedral and one trioctahedral layer. The hypothesis is put forward that both minerals formed from a dioctahedral mineral; illite from a non-expanded form and chlorite from an 'open' form by precipitation of a Fe2+/Mg hydroxide interlayer.

A method of precipitation in homogenous solution has been used for the synthesis of kaolin minerals at room temperature. The method consists essentially in producing conditions for which a sufficiently slow rate of precipitation is maintained. In the particular case described in this paper, the conditions for homogenous precipitation are obtained by addition of an anion exchange resin in the OH form to an acidic solution (pH ≈ 4) of sodium silicate and aluminium chloride or acetate.

The mineral species synthesized in each particular case depend on the kind of anions present in the original solution (chloride or acetate) and also on the nature of the exchange resin used.