hypothesis that A. xylinum is a phylogenetic ancestor of A. 8uboxydan8 thus might imply only that A. xylinum has been able to undergo consecutive loss mutations of two non-essential enzyme systems. An opposing view, namely that evolution proceeded from a simple state to a more complex one, may be more attractive on philosophical grounds. In that case, the interesting possibility has to be considered that the ancestor of Acetobacter was equipped with a mutative potential so broad as to comprise within its range two enzyme systems (respectively for cellulose synthesis and acetate oxidation) which are facultative in Acetobacter but have elsewhere assumed cardinal roles in cellular survival. SUMMARY 1. A cell-free extractofAcetobacterxyltinumfailed to form cellulose from a range of tested substrates. 2. The extract (with added phenazine methochloride) oxidized glucose aerobically to gluconate. When this system was further supplemented with diphosphopyridine nucleotide, it oxidized gluconate to oxogluconate (largely 2-oxogluconate). 3. The extract with adenosine triphosphate oxidized glucose aerobically to carbon dioxide. The extract contained kinases (glucokinase and gluconokinase), phosphoglucomutase, a complete pentosecycle set of enzymes, and a set of enzymes by which triose phosphate could be converted into pyruvate. 4. A scheme of altemate pathways for glucose metabolism that can be operated in an A. xyltinum cell is proposed. The scheme shows a pentose cycle which has several ports of entry and exit. The latter lead from the pentose cycle into a citrate cycle. 5. Biochemical evolution in the genus Acetobacter is briefly discussed. Note added in proof.