Stag Hill is an example of a slope failure in London Clay. In 1964 it was proposed to develop the area, so the slope was stabilized using a drainage system and heavy buildings were constructed on piled foundations. Piezometers were also installed to establish the pore pressure pattern. In 1985 a general evaluation of the history of the failure, stabilization and development of the area was undertaken. This gave rise to the view that this area would be a very suitable place to establish a system for studying the efficiency of drainage systems. The presence of the drainage system, which initiates rapid fluctuation of ground water pressure in response to infiltration, brought about the necessity of attention to the errors associated with time lag which would be involved in monitoring of pore water pressure. Later on, when the ground water pressure was being monitored, it was observed that changes of atmospheric pressure significantly affects the records of vibrating wire piezometers. In this way the necessity of the evaluation of difficulties and errors associated with atmospheric pressure in monitoring of ground water pressure automatically merged into the research. Accordingly, to fulfill the above mentioned aspects, all available piezometer records since 1964 were collected and plotted. All piezometers which survived were tested to see if they were still reliable. Drainage trenches were located at appropriate locations and over 60 new piezometers of different types were installed at different depths, in drained and non-drained areas, and in the trenches. Two types of pore water pressure measurements, daily and weekly, were taken. To record the duration and intensity of daily rainfall, an automatic tilting bucket system was installed in the area. To include the effect of the atmospheric pressure on the evaluation of the pore water pressure pattern, records of atmospheric pressure, taken at Gatwick airport were used. The main conclusions are that: Monitoring of ground water pressure, using stand pipe piezometers, in drained areas may lead to serious errors, say 1.0 meter head of water. Application of rigid piezometers is necessary but not enough, unless a reasonable correlation is established between piezometer reading times and rainfall periods. Changes of atmospheric pressure are a significant source of error in monitoring of ground water pressure techniques. A method to correct piezometer readings for changes in atmospheric pressure was established. At Stag Hill, the ground water pressure pattern, and in turn the stability of slopes is governed by layers of higher permeability in the range of 10[-5] - 10[-6] cm/sec. The main drainage system having a spacing to depth ratio of S/D = 3.4 - 3.8 has not efficiently improved the stability of the slopes. Surface covering (buildings and paving), however, has a dominant effect on the improvement of safety factor. At Wates House, the drainage system having S/D = 1.1, has lowered the ground water pressure with 92% efficiency. It takes some years before the ultimate efficiency of a drainage system takes effect. The efficiency of drainage trenches below invert level varies between the maximum and minimum efficiency of drains at invert level, in the long term. The theoretical methods for design of drainage trenches do not agree with practice. A practical design curve was suggested.