The objective of this project was the analysis of pattern formation in the developmental phase of the myxobacterial life cycle. Based on biological experiments and mathematical modelling we wanted to analyze, whether the cell surface-associated C-signal is sufficient for explaining various aspects of the multicellular organization of myxobacteria (in particular rippling, streaming, and aggregation) and that additional signalling mechanisms are not required (e.g. diffusive signalHng). The project allowed to establish a successful cooperation of research groups that had not existed before. Initially, the experimental group in Marburg provided experimental data related to the characterisation of individual cell motiüty. Appropriate image analysis methods and cell-based models were designed and developed by the theory groups in Dresden and Berlin. Furthermore, modelling results, particularly on the clustering of self-propelled asymmetric cells, were used to design experiments in order to validate newly generated pattern formation hypotheses. Our main efforts were directed towards the emergence of local cell ordering which is an essential characteristics of rippling, streaming, and aggregation. We designed an experimental model system (based on appropriate motility mutants) for the quantitative analysis of cell ordering. In particular, we studied a mutant with a deficiency in cell reversal and cont act-dependent motility and showed that in the vegetative phase populations of this mutant organize into compact clusters of aligned cells. The observed clustering behaviour can be explained by the hypothesis that actively unidirectionally moving rod-shaped cells may form clusters just by mechanical interaction (volume exclusion). We could link our hypothesis to the well-recognised statistical physics research field of "self-propelled particles", where we contributed novel insights to the emergence of collective migration. In particular, we showed that systems of self-propelled rods may exhibit a characteristic change in the cluster size distribution, depending on the density and/or length-to-width aspect ratio of the particles. In order to validate our theoretical results in the myxobacterial system we designed and conducted a series of clustering experiments. Imaging techniques combined with statistical data analysis allowed to extract reliable information about the intesity of local cell ordering in terms of a cluster size distribution. Especially, the theoretically predicted characteristic change in the cluster size distribution was recovered in the experiments. Thus, we have demonstrated that cell communication, e.g. mediated via C-signaling, is not required for local cell alignment, although collective migration is commonly understood as a social behaviour. Our team spent a major part of the project on cluster formation, since we had to carry out a large number of experiments and data analysis for two motility mutants (a S-motility and cell reversal deficient mutant (SA2407) and a completely non-motile mutant (SA2808)). In addition, we continued our work on rippling pattern formation. Meanwhile, our research has led to a good understanding of the rippling and the clustering patterns forming in the early developmental phase. We have only started to investigate aggregation patterns appearing subsequently. This and the later formation of three-dimensional fruiting bodies requires the development of three-dimensional models and simulations together with the design of appropriate biological experiments and threedimensional image analysis tools.