Flagellin, which constructs supercoiled filaments of the bacterial flagellum, is very difficult to crystallize because of its strong tendency to polymerize. We therefore crystallized the F41 fragment of flagellin, which does not polymerize because terminal regions that play important roles in polymerization are cleaved off. F41 was crystallized by the hanging drop vapor diffusion method in a mixture of polyethylene glycol, glycerol, and isopropanol, with a reservoir solution covered with silicon oil. The two key factors for success in growing sufficiently large crystals were isopropanol and silicon oil, which worked well to reduce the otherwise very high nucleation rate that resulted in hundreds of tiny crystals. The crystals were grown to very thin plates with thickness less than 10 microm, which made the collection of diffraction data very difficult. Freezing and annealing of the crystals and irradiation at synchrotron beamlines had to be carried out by specific methods and under specific conditions for its structure analysis at 2.0-A resolution.