A second programme of tests has been completed under commercial conditions using significantly deeper arrays of 76 mm drillholes in two rock masses, one in the Old Red Sandstone, and the other a set of basalt flows. Single packer tests were carried out as drilling proceeded using a standard mechanically-expanded packer with continuous autographic recording of flows and of input pressures. Some of the holes were also tested after drilling with a prototype double packer system instrumented with transducers to measure pressure within, above and below the test section. Results of the two methods are compared and individual test behaviour has been examined and it has been found that results were affected by changes in groundwater level produced by drilling. In any test it is unwise to assume that the effective piezometric level in the rock mass at the time and position of a test is necessarily the same as that observed in a completed drillhole.The tests have produced further practical evidence of the need to consider the scale of tests and observations in relation to the inhomogeneity of the rock mass under investigation. At these test sites there were marked variations in standing water levels, and in measured permeabilities in adjacent closely spaced drillholes.