A laboratory investigation of the thermo‐mechanical response of deep ocean sediments has been performed. The study encompasses the range of thermal loadings expected for the emplacement of a canister of radioactive waste approximately 30 m below the seafloor in deep ocean waters. The experimental program included mechanical and thermal consolidation, undrained triaxial compressive strength and constant stress level creep tests on seafloor sediments over a temperature range of 4–200 °C.

A method of computing settlement of foundations on weak subsoil deposits reinforced with granular piles is developed. Full‐scale in situ tests on skirted granular piles are carried out at four different sites consisting of loose/soft deposits. The settlement computations made by the proposed analytical procedure are compared with observed values from field tests where a good correlation is noted.

An analytical approach to membrane reinforced earth is presented. It is based on limit‐equilibrium and variational extremization. The results indicate that the potential failure surfaces are either planar or log‐spiral. The analysis utilizes a reinforcing membrane sheet that is orthogonal to the radius vector defining its intersection with the slip surface. Results of a closed‐form solution imply that: (1) The stronger the membrane the deeper the failure; (2) the membrane's elevation has little effect on the stability or on the location of the slip surface provided that failure is passing through it; (3) the presence of a membrane increases the compressive stress over the critical slip surface; and (4) the presence of a membrane decreases the soil's tensile stress that tends to develop near the crest. The results are presented in a convenient format of stability charts.

The strong influence of construction procedures was demonstrated by comprehensive field measurements of initial fining behavior and of ground movements for a soil tunnel. Plowing movements for the first tunnel shield created much of the ground loss and settlement. Much less settlement was observed for the second tunnel constructed with a shield of an improved design. Volume changes in the soil played an important role in the entire pattern of ground movement. During first tunnel construction, significant increases in soil volume resulted in less surface settlement than the actual amount of lost ground about the shield. The opposite was observed for the second shield where significant compression of soils took place, particularly in the pillar between the two tunnels and in the previously disturbed soils over the first tunnel. The rib and lagging initial lining was demonstrated to be truly flexible. A ground reaction curve was established on the basis of analysis and measurement of load. The initial lining was observed to be carrying less than half of the full overburden load.

An elementary analysis for calculation of stresses in a rammed earth house is presented. The analysis permits calculation of safe wall height, and an examination of safety of these structures. It is observed that the usual wall thickness is quite safe; the thickness can be reduced by at least a third without endangering safety and this can lead to a substantial savings in material and labor costs and an increase in the availability of space.

To investigate the flexural behavior of reinforced soil beams, tests were performed on rectangular and trapezoidal soil beams of varying span to depth ratios and varying amounts of reinforcements. Local silty clay was used in the study and simple supports for the beams were employed. The results indicate that strains are linearly distributed across the depth of the beam up to cracking and become nonlinear thereafter. Improvement due to reinforcement is smaller in trapezoidal and shallow beams. The increase in failure compressive strains is more for beams of smaller span to depth ratio and with larger amounts of reinforcements.

Various methods for the determination of the foundation model parameters are reviewed. It is then shown that one of the four known methods is conceptually of questionable validity because according to it the model parameters may be expressed analytically in terms of the elastic constants (E,υ) of the actual subgrade. As part of the presentation, the Pasternak foundation response is derived from the equations of an elastic continuum. The paper concludes with a discussion of the other methods published in the literature for the determination of these parameters.