The diffraction of x‐rays at small angles occurs in the form of sharp interferences (discontinuous) or as diffuse scattering (continuous). The first effect is observed mostly in crystallized proteins and in protein fibers. New measurements on collagen were obtained by the author in collaboration with Sekora, and independently by Bear. Using these data, an attempt has been made to establish a connection between the orders of the intense reflections belonging to the giant period of 642 Å. along the fiber axis and the frequency of the amino acid residues (frequency is defined as the number of amino acid residues of a particular kind among the total 220 residues which constitute the basic period of 642 Å. along the fiber axis). With the exception of the most abundant amino acids, glycine and proline, it is possible to assign to each frequency a corresponding reflection, and vice versa. This indicates a high degree of regularity in the arrangement of these amino acid residues. But the relationships are apparently more complex than would be expected from the Bergmann‐Niemann rule. For example, the frequencies of 11 and 20 (corresponding to an identical residue at each 220/11 or 20th location or at each 220/20 or 11th location, respectively) appear to play a role. Fundamentally, however, we arrive at a picture in accordance with the comprehensive investigations and results of Astbury. It has not yet been possible to arrive at a detailed structure.In the theory of diffuse low‐angle scattering, it is necessary to distinguish between densely packed systems and dilute systems. For the latter case, molecular scattering is computed in accordance with Debye's theory of scattering. It is shown how size factor and form factor may be separated. Both of these properties can be approximated without making previous assumptions, though the accuracy of the form factor determination can be improved appreciably if the particle weight is known. Preliminary trial measurements on solutions of edestin, tobacco mosaic virus, and hemocyanin are in agreement for the first two substances mentioned. The measurements on insulin indicate associated particles that deviate strongly from spherical shape (however, these measurements were made at an earlier date and require checking). Those on hemocyanin indicate an association similar to a string of pearls, a result which probably explains the contradiction of the substance appearing as spherical particles under the electron microscope, while exhibiting flow birefringence when in solution. When this method is developed further it will undoubtedly become a valuable tool for the determination of form and size of polymer molecules and other colloidal particles.