The paper begins with a brief overview of some of the foundation design problems for gravity structures subjected to combinations of static and cyclic loads under undrained conditions and specifies soil data needed to analyze these problems. The paper then presents results from undrained triaxial and direct simple shear tests with various combinations of static and cyclic shear stresses on Drammen clay and describes how the data needed in the analysis can be derived from these tests. The results provide general knowledge about soil behavior under cyclic loading and constitute a database which can be used to determine soil parameters for early feasibility studies of gravity structures on clay. The database can also guide in setting up laboratory testing programs for soils from actual sites and provide a framework for interpretation of the results from such testing. Further, the results in the database can be used to develop and to check the validity of theoretical models.

The paper presents a procedure to calculate the bearing capacity for foundations subjected to a combination of static and cyclic loads under undrained conditions. The term cyclic shear strength is defined, and laboratory tests are used to show how this strength and the type of failure of a soil element depend on the average and cyclic shear stresses and the number of load cycles. The calculation procedure is based on the strength determined from laboratory tests. It ensures strain compatibility of the soil elements along the potential failure surface and accounts for redistribution of average stresses during undrained cyclic loading. The procedure gives the failure load, the location of the critical failure surface, and indicates whether the foundation failure will occur as large cyclic displacements, large average displacements (e.g. settlements), or a combination of the two. The paper also presents examples with numerical values of cyclic shear strengths from triaxial and DSS laboratory tests on Drammen clay, both for given number of cycles and for typical storm loading.

A boundary‐element solution procedure is illustrated and applied to the analyses of both steady‐state and non‐steady‐state unconfined seepage problems. These analyses are performed for a laboratory model and prototype earth dams to determine the position of the free surface in a steady‐state condition, as well as the movement of the free surface in a transient condition resulting from drawdown of the water level in a reservoir. Applicability of this method is evaluated based on comparisons between the numerical solutions and measured laboratory and field data.

During the past decade, the theory for heave prediction has developed within the context of unsaturated soil behavior and has become a valuable tool for geotechnical practice. The laboratory procedures for testing expansive soils have also been essentially standardized. The heave prediction theory is briefly reviewed in this paper, and the importance of sampling disturbance is emphasized. Closed‐form solutions are presented for several possible situations that can be applied to engineering practice. In all cases, the soil deposit is assumed to be homogeneous, and the swelling pressure is assumed to be constant with depth. The closed‐form solutions are presented for the calculation of total heave when: (1) A portion or the entire “active depth” is wetted; (2) a portion of the expansive soil is excavated; and (3) the excavated portion is backfilled with a nonexpansive soil.

Erosion rates of unsaturated compacted sodium and calcium montmorillonite clays are measured under a range of tractive shear stresses. The calcium montmorillonite is treated with additives calcium chloride and sodium carbonate to obtain samples with different chemical compositions. Samples are compacted at near the optimum water content and subjected to flowing water in the flume. The tractive stress is determined by measuring the velocity distribution of the flow and using the Prandtl‐von Karman equation. The tractive stress ranges from 1.67–12.9 Pa. The erosion rate of calcium montmorillonite (a nondispersive clay) is two orders of magnitude higher than that of sodium montmorillonite (a dispersive clay). When calcium montmorillonite is treated with 0.88% by dry weight sodium carbonate, the erosion rate is reduced to the same order as sodium montmorillonite. Empirical equations are developed to estimate the erosion rate of compacted unsaturated montmorillonite clay as a function of the sodium adsorption ratio and tractive stress.

The development of the Cam‐clay model, based on critical state concepts associated with the plasticity theory, is a consequence of the geotechnical engineer's desire to simply predict the behavior of fine‐grained soils with a small number of easily measurable input parameters. Encouraged by the success in generalizing the compressibility of fine‐grained saturated soils using the Gouy‐Chapman diffuse double‐layer theory, an attempt has been made to logically extend this theory to generalize the shear strength behavior in the working stress range of engineering interest. This has indicated the possibility of generating a unique Cam‐clay model, which in turn implies a constant value of M for any soil. From basic considerations, the constancy of M has been substantiated, and the reasons for deviations from this constant value in actual practice have been indicated. It is believed that such generalizations will greatly enhance the predictive capability and practical applicability of the Cam‐clay model.

A one‐dimensional mathematical model based on finitestrain theory is developed to solve the problem of seepage‐induced consolidation in sedimented slurries or very soft clays. The direct solution employs known or assumed material property relationships to determine the final thickness of a soft sediment subjected to a constant piezometric head. It is useful for predicting the capacity of a disposal area and the time‐dependent improvement in material properties. Alternatively, the inverse solution utilizes final settlement and steady‐state flow data from laboratory or field tests to deduce permeability and compressibility relationships for soft sediments. This approach is especially helpful in the case of permeability determinations because it avoids some of the major problems associated with permeability testing of such materials. The resulting model shows that the coefficient of permeability influences both the time to reach the steady‐state condition and the nature of the steady‐state condition itself. An illustrative example is presented wherein data from a series of tests on a kaolinite slurry are used to establish material property relationships that are then used to predict the response of other tests on the same soil under different conditions.

The effect of Poisson's ratio on the disturbance index is considered by obtaining solutions to expansion of cylindrical cavities in layered elastic system. For a specific size of the remolded annulus, the disturbance index becomes lower as the disturbance becomes smaller and Poisson's ratio becomes larger, and it decreases much faster with decreasing disturbance than with increasing Poisson's ratio. Also, for a specific value of Poisson's ratio, the rate of decrease of the disturbance index with decreasing disturbance is much faster for a larger annulus than for a smaller one. In general, the effect of Poisson's ratio on the disturbance index is more pronounced for a small remolded annulus than for a large one and it is minimal for a very small disturbance. The results can be used to evaluate the modulus of deformation of the undisturbed soil mass from the pressuremeter test results and are presented in the form of graphs suitable for practical applications.