WNT signalling orchestrates a number of developmental programs 1, 2, 3. In response to this stimulus, cytoplasmic β-catenin (encoded by CTNNB1) is stabilized, enabling downstream transcriptional activation by members of the LEF/TCF family 4, 5. One of the target genes for β-catenin/TCF encodes c-MYC, explaining why constitutive activation of the WNT pathway can lead to cancer, particularly in the colon 6. Most colon cancers arise from mutations in the gene encoding adenomatous polyposis coli (APC), a protein required for ubiquitin-mediated degradation of β-catenin 7, but a small percentage of colon and some other cancers harbour β-catenin–stabilizing mutations (refs 8, 9, 10, 11, 12, 13, 14, 15, 16, 17). Recently, we discovered that transgenic mice expressing an activated β-catenin are predisposed to developing skin tumours resembling pilomatricomas 18. Given that the skin of these adult mice also exhibits signs of de novo hair-follicle morphogenesis, we wondered whether human pilomatricomas might originate from hair matrix cells and whether they might possess β-catenin–stabilizing mutations. Here, we explore the cell origin and aetiology of this common human skin tumour. We found nuclear LEF-1 in the dividing tumour cells, providing biochemical evidence that pilomatricomas are derived from hair matrix cells. At least 75% of these tumours possess mutations affecting the amino-terminal segment, normally involved in phosphorylation-dependent, ubiquitin-mediated degradation of the protein. This percentage of CTNNB1 mutations is greater than in all other human tumours examined thus far, and directly implicates β-catenin/LEF misregulation as the major cause of hair matrix cell tumorigenesis in humans.