Analogue modelling is a simplification of Nature; using this method, structures formed due to the deformation of rocks, whose in situ properties may be poorly known, can be modelled and investigated. Our back of knowledge about ＇＇real＇＇ rocks is one of the major limitations to any kind of modelling of deformation. However, unlike their counterparts in Nature, the initial (undeformed) stages of analogue models can be documented, and can easily be compared with their later deformed stages. This comparison is essential to gain an understanding of the evolution of the resulting structures. Analogue modelling is a relatively simple and inexpensive technique which can be very valuable as long as its limitations are well understood. In rock mechanics, real rocks are strained over human time scales. As many of the properties of rocks are time-dependent, the results of this type of experiment cannot directly be extrapolated to geologic time scales. Instead, modellers diminish the temporal and spatial dimensions by substituting carefully-chosen analogue materials for real rocks. Unlike field geologists, analogue modellers are able to follow the development of their models in a fixed reference frame. Cloos (1955) noted that: ＇＇Many interpretations would never have been published if the author had only once tried his suggested mechanism of folding or faulting in an experiment....＇＇; he also concluded that ＇＇[analogue] experimenting is a good deal of fun ＇＇!