Potassium incorporation in the structure of combeite has been studied in detail. Since natural combeites are known to contain only small amounts of potassium focus was laid on the Na-rich part of a hypothetical solid-solution series with composition Na2-xKxCa2Si3O9. Samples were prepared from mixtures of silica and the corresponding carbonates for nominal compositions with x = 0.2, 0.3 and 0.5, heated from ambient temperature to 1350 °C and slowly cooled to 1000 °C. After disintegration of the carbonates, the platinum capsules used as sample containers were welded shut in order to avoid losses of the volatile K2O and Na2O components. From all three batches potassium containing combeite crystals could be retrieved. Single-crystal diffraction experiments revealed the following compositions: Na2.10(1)K0.11(1)Ca1.90(1)Si3O9, Na2.09(1)K0.18(1)Ca1.91(1)Si3O9 and Na2.13(1)K0.18(1)Ca1.87(1)Si3O9. Consistently, the trigonal crystals (space group P 31 2 1) contained (i) (K + Na):Ca ratios larger than 1:1 and (ii) potassium concentrations lower than those in the starting mixtures. Since the K-contents of the samples obtained from the runs with x = 0.3 and 0.5 were almost identical, the solid-solution seems to be rather limited with an upper boundary of about one potassium atom per unit cell. The structure of the K-containing combeites is very close to the K-free structures reported in the literature. It can be described as a mixed tetrahedral-octahedral network in which additional K, Na and Ca cations are incorporated for charge compensation. A detailed analysis of the topological features of the net is presented. From the six observed extra-framework sites only the M22 position showing a coordination environment with ten next oxygen neighbours is involved in the K-substitution. Potassium uptake is also reflected in increasing values for the lattice parameters a and c as well as the unit-cell volumes. Actually, the c-axis is more affected from the incorporation of the comparatively large K+-cations.