Previous thread rolling methods for the plastic limit are shown to be inadequate and inaccurate. Alternative methods for the plastic limit are shown to be imprecise and unreliable. The strength-based concept and use of the fall-cone test to determine the plastic limit are shown to be flawed. An apparatus that replicates Atterberg’s rolling technique, devised and developed by the author, is described, referred to as the Barnes Apparatus. A thread of soil is rolled between two plates configured to permit extrusion and reduction of diameter with much less operator interference than with the standard test and judgement of the crumbling condition is eliminated. Using a loading device nominal stresses are derived and from dial gauge readings diametral strains are determined for each rolling traverse of the soil thread. Toughness has previously only been studied in an empirical or qualitative manner. From plots of nominal stress vs. strain the workability or toughness of the plastic soil is determined as the work/unit volume. The apparatus and test are appropriate to a wide range of soils. Threads are tested over a range of water contents from near the sticky limit to the brittle state. Good correlations between toughness and water content display an abrupt ductile-brittle transition and give an accurate definition of the plastic limit. From the correlations useful properties are obtained such as the maximum toughness at the plastic limit, the toughness limit, the water content at zero toughness, the stiffness transition, the toughness coefficients, the toughness index and the workability index. An investigation into the significance of the soil thread diameter of 3 mm in the standard plastic limit test has found that as the water content of a soil reduces it undergoes a transition from fully plastic, to cracked, to brittle, largely regardless of the diameter of the thread. It is recommended that the 3 mm diameter requirement is withdrawn from the standard test procedure as unnecessary and emphasis placed on observing the behaviour of the soil thread as it is rolled by hand. A review of the relationship between the clay matrix and the granular particles in a soil has found that the linear law of mixtures and activity index are appropriate only at high clay contents. The terms granular spacing ratio and matrix porosity are introduced to explain the effect of the granular particles on the toughness and plastic limit. An analysis confirms that with small diameter soil threads large granular particles affect the results disproportionately. An aggregation ratio term is introduced to explain the change in toughness in the clay matrix as its water content reduces towards the plastic limit. To assess the effect of granular particles in a clay matrix on the toughness and plastic limit the results of tests conducted on mixtures of a high plasticity clay and silt, and sand particles of two different sizes are discussed. Smaller particle sizes are found to have a greater effect on reducing the toughness and the plastic limit of the clay. In the ceramics industry mixing different clays together to obtain suitable properties is common. The toughness and plastic limits of two pairs of mixed clays do not follow the linear law of mixtures but are dependent on the total clay content and the content of a dominant clay mineral.