The suggestion that the amylose molecule has a helical structure is certainly one of the most interesting that has been put forth in connection with the study of starch. Enzymatic studies have led many investigators to the conclusion that the starch molecule is constructed in a manner that would permit a unit of six glucose molecules to be split out during degradation in order to explain the preponderance of hexasaccharides that are found after the limited treatment of starch with -amylase from malt (20). The formation of cyclohexaamylose and cycloheptaamylose by the enzyme of bacillus macerans lends credence to the idea that the starch chain is helical, with six or seven glucose units per turn (10). In 1939, Freudenberg and coworkers (11) constructed a model of a linear starch molecule from Stuart atomic models and concluded that the helical structure was extremely plausible. Six glucose residues could be fitted into one turn without strain and produced a helix with a diameter of 13 A. and a length of SUMMARY 1. From sedimentation-velocity, diffusion, and viscosity measurements on very homogeneous samples of amylose triacetates, the dimensions of the ellipsoids of revolution which are hydrodynamically equivalent to the shape of the molecules are calculated.2. The resulting dimensions are compatible with the helical model of amylose which has been established for the solid amylose-iodine complex by x-ray methods.The authors are greatly indebted to Dr. Ralph W. Kerr of the Corn Products Refining Company, Argo, Illinois, for the samples of crystalline amyloses and to the Corn Industries Research Foundation for a gift that in part supported this research.