Dispersive soils are prevalent in many areas of South Africa and the presence of these soils has not only posed a problem in earth dam construction but has led to problems with a number of road projects. The use of dispersive soils in roadway embankments and structures can lead to serious engineering problems if the soils are not accurately identified before use and appropriate mitigation measures taken. Although the causes and consequences of soil dispersion are well understood, the consistent and positive identification of dispersive soils still remains a problem. Many identification methods have been proposed but none has been completely successful. The current tests used for identification include the pinhole, double hydrometer, crumb and chemical tests, which are generally used in combination to obtain the most reliable outcome. These laboratory tests, however, have not always been entirely consistent, either when used in combination or individually, and it is possible that the reason lies in the actual testing procedures. Recent investigations have shown that in many cases the identification and classification problems appear to be related to inconsistencies in the test methods and testing protocols. This has been highlighted recently at various conferences and presentations in South Africa, with a recommendation that the problem be investigated fully. The main objective of this project was thus to carry out a detailed investigation into the current methods used for the testing and identification of dispersive soils. The test methods were thoroughly analysed and shortcomings identified. The differences in results and interpretation resulting from different test techniques are examined and solutions to overcome the problems proposed. A further objective of this study is the assessment of the interpretation of results obtained from dispersive soil chemical analysis. A pilot study was first carried out, which recognized deficiencies in the identification process resulting from problems with the test methods. The pilot study allowed for the identification of these shortcomings in the test methods and their modification. The modified methods were used in the full study to obtain more consistent and reliable results. The study found the pinhole test as well as the Gerber and Harmse (1987) ESP versus CEC chart to be highly unreliable. The double hydrometer and crumb tests are good indicators of dispersivity if carried out accurately and repeatably using a standard test method. The chemical analysis of the soils should be carried out using the standard methods employed by the Soil Science Society of South Africa. Final rating systems were also found to be unreliable since they put a great deal of weight on the poorly reproducible pinhole test. Emphasis is thus put on the test methods being as simple and unambiguous as possible to promote repeatability and reproducibility of the results. Recommendations are finally made proposing a suite of tests as well as a decision process which should be followed when faced with a potentially dispersive soil. This investigation was aimed solely at the evaluation of test protocols and not the suitability or specification of limits for any rating systems. This should now be assessed using the standard test methods proposed in this thesis. / Dissertation (MSc)--University of Pretoria, 2013. / gm2014 / Geology / unrestricted