Twelve papers featured in a special issue on individual-based modelling in ecology are reviewed in an effort to identify common methodological and theoretical issues. The review focuses on issues related to the question of whether and how individual-based modelling is changing ecological theory. One major hindrance impeding the generation of theory from individual-based models (IBMs) is the fact that IBMs are more or less complex computer simulation models. They are thus hard to develop, hard to communicate, and hard to analyse. Solving this problem requires both software tools which help to implement and communicate IBMs and at least the same effort in analysing the models as is currently put into their development. A new field of application of IBMs is Virtual Ecology, i.e. the comparison of simulated data sets with those obtained by virtual (i.e. simulated) ecologists. This method allows field methods, empirical measures and sampling protocols to be optimised. As far as theoretical issues are concerned, individual variability was by far the most important issue discussed in the papers. Previously most studies concentrated on the mechanisms generating individual variability, but there is now also a growing number of models addressing the consequences of individual variability for population and community dynamics. In order to determine these consequences, a currency is required which allows the model populations to be evaluated. Persistence and other stability properties are proposed as such a unifying currency. The lesson contained in our review is that even with just twelve papers more or less explicitly oriented towards ecological theory, elements of a theory that might emerge from individual-based modelling in the future can already be identified.