David Wilkinson's review in this issue reminds us that pattern formation is a central concept to the study of neural development. While empirical observations have begun to elucidate patterning mechanisms, the theoretical analysis of pattern formation also has a long tradition. Hans Meinhardt and Alfred Gierer have advocated this approach since the early 1970s, and they have collected an impressive array of mathematical models to support their ideas. These models can be found on their web site, 'Theoretical aspects of pattern formation and neuronal development'. The site, in Meinhardt's words, “. . . is addressed primarily for experimental scientists, hoping to convince them that an integration of the theoretical and the experimental approach can provide a deeper understanding of what is going on. Emphasis is placed on the self-organizing aspect of development and on the generation of patterns starting from near-uniform tissues and cells”.

The site uses animated computer simulations to illustrate a variety of aspects of pattern formation, from the establishment of the embryonic axes to the pigmentation of the mollusc shell. There is plenty to interest the neuroscientist, with sections on axon guidance, growth cone navigation and the roles of boundaries as organizing regions. Admittedly, some of the animations are hard to follow because they run very fast, and they would benefit from some additional labelling, but otherwise this is a very useful resource.

The site is clearly written, and provides an accessible introduction to an aspect of developmental biology that scientists often dismiss as being too 'difficult'. Meinhardt recognizes that many biologists have an aversion to mathematics, and he hopes that this web site will help to encourage them to join in with his theoretical approach.