Present-day understanding of the architecture of the entire cell-wall of Mycobacterium tuberculosis amounts to a "core" template comprised of peptidoglycan with phosphodiester linkage, via a linker disaccharide, to a linear D-galactofuran, to which, in turn, are attached several strands of a highly branched D-arabinofuran. The cell-wall mycolic acids are linked via an ester to the majority of the non-reducing termini of the D-arabinan. The mycolic acids are oriented perpendicular to the plane of the membrane and provide a truly special lipid barrier responsible for many of the physiological and disease-inducing aspects of M. tuberculosis. Intercalated within this environment are the phthiocerol dimycocerosates, cord factor or dimycolyltrehalose, sulfolipids, phosphatidylinositol mannosides and the related lipomannan and lipoarabinomannan, etc., agents responsible for much of the pathogenesis of tuberculosis. Interest in the biosynthesis of the cell-wall core, regarded, unlike the ancillary lipids, as essential to bacterial viability and integrity, is now driven by the pressing need for alternative drugs to counteract drug-resistant tuberculosis. In a manner analogous to the roles of lipid I and II in peptidglycan formation, synthesis of the entire arabinogalactan is initiated by transferring activated sugars to decaprenyl-phosphate, giving rise to the linker disaccharide, followed by stepwise elongation of the galactan, and the arabinan, apparently one sugar at a time. The genetics and enzymology of these polymerization events have not been well defined, nor have the final steps, namely the attachment of mycolic acids and ligation to peptidoglycan. However, what is known of the earlier events in cell-wall core synthesis has attracted interest in terms of new anti-tuberculosis drug development.