Little is known about the clonal structure or stem cell architecture of the human small intestinal crypt/villus unit, or how mutations spread and become fixed. Using mitochondrial DNA (mtDNA) mutations as a marker of clonal expansion of stem cell progeny, we aimed to provide answers to these questions. Enzyme histochemistry (for cytochrome c oxidase and succinate dehydrogenase) was performed on frozen sections of normal human duodenum. Laser-capture microdissected cells were taken from crypts/villi. The entire mitochondrial genome was amplified using a nested PCR protocol; sequencing identified mutations and immunohistochemistry demonstrated specific cell lineages. Cytochrome c oxidase-deficient small bowel crypts were observed within all sections: negative crypts contained the same clonal mutation and all differentiated epithelial lineages were present, indicating a common stem cell origin. Mixed crypts were also detected, confirming the existence of multiple stem cells. We observed crypts where Paneth cells were positive but the rest of the crypt was deficient. We have demonstrated patches of deficient crypts that shared a common mutation, suggesting that they have divided by fission. We have shown that all cells within a small intestinal crypt are derived from one common stem cell. Partially-mutated crypts revealed some novel features of Paneth cell biology, suggesting that either they are long-lived or a committed Paneth cell-specific long-lived progenitor was present. We have demonstrated that mutations are fixed in the small bowel by fission and this has important implications for adenoma development.