Loss of ⁴⁰Ar* from plutonic alkali feldspars over geological time, and the shape of ⁴⁰Ar/³⁹Ar step-heating spectra, depend on intracrystal microtextures that provide ‘fast-pathways’ for Ar diffusion. Pathways may develop during initial cooling or during subsequent replacement events, and have their own ‘thermochronology’. Few studies have combined Ar analysis with EM, and understanding of microtextures has developed at a slow pace compared with ⁴⁰Ar/³⁹Ar thermochronometry. Qualitative chronologies can be deduced for most microtextures. Fully coherent µm-scale perthitic exsolution textures usually form first, by volume diffusion. Estimates of cooling rate can be made. They retain Ar but in granites become semicoherent, developing misfit dislocations. Their role in Ar studies needs to be explored. Irregular patch and vein intergrowths cross-cut early coherent intergrowths and form by dissolution–reprecipitation reactions in fluids. The intergrowths are on scales up to 100s of µm, but the coexisting phases are composed of µm Ab- and Or-rich subgrains with micropores at their boundaries. Pores and subgrain boundaries can act as both pathways and traps for Ar. In the Klokken intrusion this replacement was at ~500°C. Or-rich subgrains then underwent a second phase of exsolution by volume diffusion to form semi-coherent intergrowths. A second phase of replacement by very porous, structurally complex feldspar occurred at <170°C. In principle multiple phases of replacement can occur over geological time.

Most Ar extraction during step heating occurs at T well above those experienced by the crystal during its geological history and pathways are inevitably modified. Homogenization of coherent lamellae and subgrains by K–Na interdiffusion begins as soon as heating starts. A µm coherent lamella will homogenize in <1 s at 1100°C, but twinning persists because it depends on slow Si–Al interdiffusion. The instantaneous shearing triclinic–monoclinic symmetry change is unlikely to affect Ar diffusivity. Fluid inclusions move and defects related to semicoherent film lamellae begin to change their character below 500°C. A generic model for natural and laboratory Ar loss from complex alkali feldspars is unattainable. Routine combination of EM and Ar work, could, in contrast, prove extremely rewarding.