Aflatoxins are potent mycotoxins and carcinogens produced by strains of Aspergillus flavus and A. parasiticus growing on stored field crops. Most of what is known about aflatoxin biosynthesis has been learned from labeled precursor feeding experiments and from isolating and refeeding intermediates from blocked producer strains. Only recently has the characterization of a few of the biosynthetic enzymes been initiated. Aflatoxin biosynthesis begins with the formation of an enzyme-bound polyketide, produced by condensation and decarboxylation of acetyl- and malonyl-coenzyme A units. The polyketide undergoes a series of cyclization and aromatization reactions to produce a C-20 anthrone, which is released from the enzyme and oxidized to norsolinic acid. A series of reduction, oxidation and cyclization steps convert norsolinic acid through averantin and averufin to the well-established intermediate versiconal hemiacetal acetate. It is converted to the mycotoxin sterigmatocystin by way of versicolorin A hemiacetal and versicolorin A. Conversion of versicolorin A to sterigmatocystin requires a series of steps including oxidative ring opening, oxidative coupling, reduction and rearrangement with loss of water and carbon dioxide. Conversion of sterigmatocystin to aflatoxin B1 has been studied as an enzymatic process in several types of cell-free preparations. The process involves methylation by a soluble enzyme and a sequence of oxidative ring cleavage, condensation and decarboxylation reactions catalyzed by membrane-associated enzyme(s). Formation of aflatoxin B2 and other aflatoxins appears to involve similar conversions of the corresponding sterigmatocystin analogs rather than metabolism of aflatoxin B1.