Anisotropy of clays was studied from cuboidal shear tests on two different clays that were artificially sedimented in a slurry consolidometer. These were supplemented by consolidation tests on specimens cut at different orientations. For one‐dimensionally consolidated specimens, a significant increase was observed in the angle of internal friction when the specimens were loaded horizontally. For isotropically consolidated specimens, the angle of internal friction was about the same for vertical and horizontal loading. From the limited data available, it appears that the difference increases with the inherent anisotropy of the clay fabric. It was also found that when the one‐dimensionally consolidated clay specimen is loaded horizontally, the rotation of principal stresses takes place. This results in the development of very high pore pressures at failure. The substantial increase in the angle of internal friction and the development of very high pore pressures at failure for horizontal loading of one‐dimensionally consolidated clays are very important considerations in the geotechnical problems in which the soil is loaded horizontally.

This paper describes a simplified hybrid layer approach for settlement analysis of axially loaded pile groups. The shaft and base interactions are considered separately in individual piles in a group. Semianalytical closed‐form expressions are derived to evaluate the shaft and base flexibility coefficients. Both linear and nonlinear pile responses are considered, and the soil mass may be arbitrarily layered. Good agreement is obtained between theoretical solutions computed by use of the approach and the more rigorous boundary integral approach. Finally, the approach is used to compute the settlement response of two field pile groups embedded in stiff overconsolidated clays, and reasonable agreement is observed between the measured and computed behavior.

Seismic cone penetration testing (SCPT), a relatively new in situ technique, records the arrival of seismic waves generated at the surface using velocity or acceleration transducers installed in an electric piezocone. To distinguish the different seismic arrivals, instruments with fast response times are required, i.e., accelerometers. The fast response time of an accelerometer results in a very sensitive instrument with corresponding noisy time‐domain characteristics. One way to separate events is to apply digital filters, which characterize the signal frequencies and then isolate frequency spectrums of interest. The digital filter introduced in this paper is based on frequency‐domain filtering using cross‐correlation algorithms. The cross‐correlation filter concepts are synthesized by a computer program referred to as PS CrossCor, whose performance was evaluated by interpreting both synthetic and real data. Data from three sites is presented to compare the PS CrossCor and the established crossover method. The results from the analysis indicate that PS CrossCor is an accurate and reliable tool that greatly assists in determining seismic velocities, and has potential for determining other seismic parameters.

A new method for the study of soil fabric has been developed. This method has been referred to as conductometric phase transition porosimetry (CPTP). With CPTP, changes in electrical conductance and temperature of saturated soil samples are measured during a cycle of capillary freezing and melting. These measurements are used to calculate the sample's pore size distribution. The paper includes an introduction to CPTP, an investigation of the variability of test results, a comparison of results with mercury intrusion porosimetry (MIP), and a discussion of the advantages and disadvantages of CPTP. Experimental results show that the new methodology can provide a useful and valuable tool for quantitative studies of soil fabric and is capable of measuring pore sizes between 2.5 nm and 5,000 nm diameter. Pore sizes measured by CPTP are smaller than MIP‐measured pore sizes. Due to the lesser disturbance of pore structure and much larger sample size, CPTP provides a more accurate and representative measure of the pore structure of recompacted fine‐grained soils than MIP.

A method involving the use of driven‐rod test in conjunction with the measurement of high‐strain test (HST) data is introduced as an in‐situ testing technique for the determination of Smith model parameters, including the static soil resistance to pile driving and its distribution, the soil quake, and the Smith damping factor. A total of nine pile driving sites were visited to gather HST data for both prototype pile driving and dynamic small‐rod probing. Subsequently, a wave equation‐based computer program CAPWAP was used to derive the Smith model parameters following the CAPWAP signal match procedure. The model parameters deduced from HST data of the 30.5 cm outside diameter (OD) pipe piles and from the 2.8 cm OD driven rods exhibit significant differences. The observed differences are attributed to the differences in the shaft sizes, the rate of penetration (both velocity and acceleration), and possibly boundary conditions. An empirical extrapolation law based on the correlation studies is proposed for transferring the driven‐rod deduced Smith model parameters to the prototype pile driving conditions. To shed further insight on the observed differences of model parameters, examinations by means of cavity expansion theories are forwarded. The first‐order estimate of the effects of the size and rate of a penetrating shaft conforms with those derived from field HST data and the CAPWAP analysis. The proposed in‐situ testing techniques offers a viable method for the determination of Smith model parameters in an initial geotechnical site investigation, and can improve the reliability of wave equation‐predicted bearing graph for pile installation control.

Results of resonant column tests were used to determine the effects of overconsolidation ratio (OCR) and of aging on the normalized rate of secondary increase $N_G$ of low‐amplitude shear modulus $G_0$ of a remolded kaolinite clay. The value of $N_G$ decreased approximately linearly with the logarithm of OCR and with the logarithm of duration of aging of the clay. The similarity of behavior provided a basis for establishing an equivalency between age‐and‐stress‐induced equivalent overconsolidation for the clay in question. It was concluded that up to a certain limiting duration of aging the age‐induced OCR increases linearly with elapsed time, whereas the effect starts diminishing for longer durations of aging. It is suggested that the results of similar studies on natural clay soils be used for predicting the long‐term behavior of clays from the results of short‐duration tests.

A common procedure to determine the expansion potential of a soil is to use the Uniform Building Code expansion index test. The soil sample is compacted into a confining ring using a specified compaction energy. If the degree of saturation is between 49% and 51%, the specimen is ready for testing, or else the test specimen procedure is repeated after adjustment of initial water content. The Uniform Building Code expansion index test attempts to control all variables that influence the expansive characteristics of a particular soil. Yet soils having low specific gravity and clayey gravels may not be accurately classified. This is because the test specifications require the use of a specific gravity of 2.7 and there is no correction to account for particles retained on the No. 4 sieve. The expansion index should also be calculated at the end of primary swell, rather than at the time indicated in the test specifications. The weighted expansion index is used for nonuniform soils, but virtually ignores the expansiveness of the soil below a depth of 0.91 m (3 ft).