The dynamic response of a 10.7 m (35 ft) thick dry sand layer was modeled on the Cambridge University centrifuge at scale factors of 35 and 80. A stack of teflon‐coated aluminum rings and a latex membrane confined a cylindrically shaped sand model that had a height‐to‐diameter ratio equal to 1. During the scaled sinusoidal base shaking, electronic transducers measured accelerations, transient horizontal displacements and surface settlements. While sensors measuring accelerations and surface settlements worked well, the DC LVDT's used to measure transient horizontal displacements were influenced by frequency‐dependent behavior. Measured results were used to evaluate scaling laws governing dynamic centrifugal modeling and to test for consistency with results of numerical methods and laboratory tests reported in the literature. The results of the test program indicate that a simple sand layer can be effectively modeled aboard a centrifuge and that horizontal accelerations, cyclic shear strains, and settlements follow the scaling laws that govern dynamic centrifugal modeling.

A simple test applicable to the evaluation of soils for water‐retaining soil structures is described. It combines into one test an appraisal for dispersivity, cracking, softening or slaking in the presence of water, measuring at the same time the influence on the soil of divergence from optimum compaction conditions. The latter aspect has particular value for the selection of compaction limits in construction specifications. The test proposed requires only modest quantities of soil, and is thus particularly suitable with variable soils, for which numerous tests are desirable.

Soil arching action or “bin effect” is usually quantified by use of a horizontal differential element whose support derives in part from Rankine theory. In the 1940's, Krynine mathematically proved this incorrect. The present analysis substitutes a catenary arch describing the path of the minor principal stress, which thus is complementary to a structural arch, and dips downward instead of upward if supportive. Soil arching action develops in two stages: The first involves rotation of the principal stresses adjacent to a rough wall and causes horizontal wall pressures to significantly exceed those from classical theory simulating a $K_o$ pressure distribution even in loose backfill soil. The second stage reduces pressures on the lower wall to give a curvilinear distribution typically centered at a height 0.42 times the height of the wall and in close agreement with published data.

A method is proposed to predict the behavior of single piles in cohesive soil subjected to vertical loads applied at various rates. The elementary simple shear model is made rate‐dependent by introducing the time raised to a negative power n. This viscous exponent n is shown to vary from 0.02–0.08 with a 0.05 average. Correlations with undrained shear strength, water content, plasticity, the liquidity index and overconsolidation ratio are shown based on 152 laboratory tests. Integration of the elementary model leads to rate‐dependent elastic‐plastic or hyperbolic load transfer curves. A program to predict the head load‐head movement response of the pile using these rate‐dependent load‐transfer curves is described. The validity of the method is first addressed by examining the physical reasons for the rate‐dependent properties of clays, then by using the results of 62 rate‐dependent pile load tests and by examining a series of applications of the new method and, finally, by presenting the results of a parametric study.

Stability analyses for stage construction of a breakwater located offshore New Jersey required a detailed evaluation of the undrained stress‐strain‐strength properties of a plastic marine clay. One set of analyses, representing “conventional” practice, used UU triaxial compression tests to obtain the initial undrained strength $\left(c_u\right)$ and CIU triaxial compression tests to predict strength increases with consolidation. The second approach employed the Stress History And Normalized Soil Engineering Properties (SHANSEP) design procedure. This involved extensive consolidation testing to measure the in‐situ preconsolidation pressure and a comprehensive program of ${\mathrm{CK}}_{o}U$ tests sheared in compression, extension and direct simple shear modes of failure at varying overconsolidation ratios. A strain compatibility technique was applied to the SHANSEP data to account for the effects of anisotropy and progressive failure in circular arc‐ and wedge‐type stability analyses. The SHANSEP approach yielded design strengths significantly smaller than conventional practice. In terms of both the initial $c_u$ profile and the rate‐of‐strength gain with consolidation.

For ease and economy, most commercial laboratories perform consolidated undrained triaxial shear tests using an initial isotropic state of stress. The insitu state of stress for most clay soils, however, is anisotropic. How different are the undrained shear strengths and effective stress friction angles measured under “field conditions” (CAU) as opposed to routine laboratory conditions (CIU)? This study reviews available published data from over 40 different clays consolidated under both isotropic and anisotropic conditions before triaxial shear. Most of these clays were normally‐consolidated, although onethird were also tested at overconsolidated states. Important factors such as strain rates, soil structure, and differences in laboratories were considered to be beyond the scope of this study. For triaxial compression, it is suggested that the anisotropic undrained srength, on the average, may be estimated as 87% of the isotropic strength. Based on few data, a tentative correction factor of 0.60 is recommended for isotropic extension to account for stress anisotropy. The effective stress friction angle appears little affected by initial stress state.

A method is presented for determining the axial static pile capacity from dynamic measurements of force and acceleration made under the impact of a large hammer. The basic equation for calculation of the forces resisting pile penetration is derived. The limitations of the basic resistance equation are discussed and illustrative examples of field measurements are given. With the availability of this derivation, it is possible to prove that the Case Pile Wave Analysis Program (CAPWAP) resistance force distribution is unique. Using the assumption that the resistance to penetration can be divided into static and dynamic parts, an expression is developed for calculating the dynamic, resistance to penetration. The resulting method requires the selection of a damping constant which is shown empirically, to relate to soil size distribution. A correlation of Case Method capacity and the capacity observed in static logd tests is given for 69 statically tested piles that were also tested dynamically.

Existing methods for the evaluation of liquefaction potential using data obtained from the Cone Penetration Test (CPT) are reviewed. A modified CPT‐based method for the liquefaction assessment of sand is presented. Information available on the liquefaction of silty sands is reviewed, and a procedure for considering the influence of silt content is also presented. Field and laboratory data from Canada, Japan, China and the United States are presented to provide a preliminary evaluation of the proposed CPT‐based liquefaction assessment methods.

A simple and realistic analytical procedure is described to estimate the eveloped lateral earth pressure behind the rigid retaining wall with cohesionless backfill soil experiencing outward tilt about the base. Included are various stages of wall tilt starting from an initial active state to a full active state. The initial active state is defined as a stage of wall tilt when the soil element at the ground surface experiences a sufficient amount of lateral displacement to reach an active condition, whereas the full active state occurs when the soil elements along the entire depth of the wall are in active condition. The predictions from the developed method of analysis are compared with the model test measurements. The comparison shows very good agreements at various stages of the retaining wall tilt.

The relative safety of neighboring geotechnical structures depends on the degree of dependence between them. This dependence is considered to exist if there is overlapping of their potential failure surfaces. A simple method is presented which defines the interference between two geotechnical structures in terms of a non-dimensional factor α, the distance needed between them to satisfy a certain confidence level that they are independent is obtained.