Recently we reported on the structures 1 and g of two degradation products of mocimycin (1). Further work has now led to the elucidation of the structure of the complete mocimycin molecule 5. The previous experiments indicated that fragment g might be an artefact and that methylation of a hydroxylgroup might have taken place under the applied conditions (acidic methanolysis). This was confirmed by treating mocimycin with acetic acid at room temperature for several days: An apolar diene, z, was obtained, together with a polar compound s, which was acetylated with a mixture of acetic anhydride and pyridine to give s. The W spectrum of 2 showed the presence of fragment 1. The structure of e was derived from its 220 MHz proton NMR spectrum. Correspondingly, acetic acid treatment of the mono 4-bromobenzylderivative of mocimycin yielded C. The spectra of s are similar to those of 3 but indicate the presence of hydroxyl groups at C-15 and C-16 (15-H and 16-H shifted to δ CDCl₃ 4.30 and 4.14 respectively; no estercarbonyl absorption present at 1735 cm⁻¹) and a free amine group (25-H shifted to δ 3.3). In mocimycin, fragments g and sa appear to be linked by an amide bond as the NMR-spectrum shows an extra amide hydrogen signal just like the compounds 2 and s, while the IR spectrum indicates the absence of a γ-lactone (no carbonyl absorption at 1785 cm⁻¹ as present in diene 2b) and the enolic hydroxyl of the pyridone fragment fully accounts for the mono-acidic character of mocimycin. As the NMR-spectrum of mocimycin does not indicate the presence of other protons than those found in 2a and 4a, structure 1 is suggested as the two-dimensional structure of mocimycin. Support for this structure was obtained when mocimycin was treated with sodium metaperiodate. This resulted in the formation of an aldehyde, which was purified as its 2,4-dinitrophenylhydrazone, 2. It incorporates both the diene fragment s and a part of 4a, as was evident from the spectral data. The suggested structure 1 explains the occurrence of mocimycin as an equilibrium mixture of three components, as shown by two-dimensional TLC (2). Two of these components, including the main one, are acids and show a positive ferrichloride reaction, whereas the third component is a neutral, ferrichloride negative substance. The configuration at the two trisubstituted double bonds (Δ⁸ and Δ¹⁰) is still uncertain, but in structures 4, 1 and z we depicted the presumably more stable trans configuration. In a solution of mocimycin in acidified aqueous acetone an acetonide is formed at C 15 and C 16, suggesting a cis relation between the two hydroxyl groups. The NMR-spectrum of z in DMSO is in fair agreement with the data published by Maehr et al. (3) for goldinono-1,4-lactone-3,7-hemiketal, which suggests that both substances are stereochemically identical also. From the strong resemblance of the NMR-spectrum of antibiotic X-5108 (4) to that of mocimycin, it would appear that antibiotic X-5108 is the pyridone N-methylated derivative of mocimycin.