Abstract The present report covers the first year of investigations carried on by the Committee at Union College, Schenectady, New York. The purpose of these investigations is to examine more closely than heretofore the fundamental characteristics of spring action in order that more reliable constants for a code of design on mechanical springs (see Bibliography, No. 18) might ultimately be established. The experiments were conducted by Anthony Hoadley under the direct supervision of M. F. Sayre. Certain of the results were covered in Progress Report No. 3, which report was not published. It should be considered as a progress report, in that, while a number of valuable results have been obtained, the field opened up is so wide that much of the work must be considered as preliminary only. Particular points of interest are: (1) A study of distribution of stresses in bending in spring steel, at the very high unit stresses often used in practice. This has practical importance in furnishing a partial explanation of the large discrepancy between theory and practice as to allowable unit stresses, in furnishing a theoretical guide as to the amount of recoil to be expected when winding a spring about a mandrel of a given size, and in determining the effect of the initial stresses induced in a spring during the winding operation, upon the future behavior of the spring. (2) A comparison of proportional limit and of modulus of elasticity for the same material in bending and in tension at high and at low stresses, both in virgin condition and after a cyclical state has been reached. This has a possible importance in determining what weight should be attached to tensile tests in measuring the suitability of a proposed spring material. It also was done in order to investigate certain theoretical and practical discrepancies between properties in bending and tension. (3) A study of the comparative internal friction in spring material in bending and in tension, for varying stress ranges, up to very high loads. This has a bearing in its effect on the possible accuracy of spring-controlled measuring instruments, and a smaller bearing on the energy losses in power springs. It is also highly desirable to know accurately how this internal friction or hysteresis varies with the load range and the position of the load range in order to determine what relation may exist between it and both the proportional limit and the fatigue limit of the material, as governing allowable safe loads. An unexpected result of the investigation is that the hysteresis in tension was found to be far less in proportion to that in bending than was expected. This result should be considered as tentative only. Every effort is being made to discover any instrumental defects which may have contributed to this result, and lacking this, to find satisfactory theoretical reasons.