AbstractObservations on the x‐ray scattering by ball‐milled cellulose and cellulose derivatives, amorphous oligosaccharides, and Fortisan H are reported and an approximate shape is established for the scattering curve of the noncrystalline component of celluloses. Using this shape measurements have been made of the minimum crystallinity which is consistent with the observed x‐ray diagram of regenerated cellulose.From a comparison of this minimum crystallinity with crystallinity determined by the infrared‐deuteration method, it is concluded that at least 14% of the total material cannot be truly described as either perfectly crystalline or perfectly amorphous. This fraction of material of intermediate order is registered as crystalline by x‐ray diffraction, but infrared spectroscopy suggests that it does not possess the precise molecular configuration characteristic of crystalline material. Support for this view of structure is reported from observations of differences in hydrogen‐bonding in amorphous regions of regenerated and bacterial celluloses. Infrared results show that OH groups which lie in the surface of crystallites are hydrogen‐bonded in a random amorphous manner and it is concluded that the material of intermediate order can be accounted for in terms of chains lying in the surface of crystallites of cross‐section 29 A. × 65 A.