An earthquake prediction model which is no longer considered to be 'in fashion'.

Developed in the 1970's, the dilatancy-diffusion theory of earthquake prediction is based on fluid pressure and microcracks within the supporting material. Dilatancy is a term for the increase in the volume of a rock during deformation: dilatancy can occur via an increase in pore volume, a change in the crack and pore distribution, the rotation of grains or microfracturing. Dilatancy is important in earthquake prediction because it changes the ratio of P wave to S wave velocity, ground level and ground water level.

The theory is as follows:
An external or internal elastic strain causes rocks to dilate. The rocks increase in volume when the stress = 50% of the rock strength. Microcracks form and increase in number, allowing more water to saturate the rock and forming even more microcracks. This lowers the rock strength and eventually leads to failure: rupture occurs, and fluid pressure and the stress on the rocks is released.

The main proponents of the theory stated that by charting the time and magnitude of these precursors, a good estimate of the time, location and magnitude of the impending earthquake could be developed. Unfortunately the precursors proved to be difficult to measure accurately and quickly enough for any sort of reliable estimate to be given: hence the theory has fallen out of favour in geophysics circles.