The dewatering of oil sands sludge is a major technological, economical, and environmental challenge to the oil sands industry of northeastern Alberta. Sludge is a mixture of small mineral particles (less than 44 µm in diameter), residual bitumen from the extraction process, and water. Sludge consolidates at the bottom of tailings ponds to approximately 30% solids in 2 years and will remain at this level of solids and water indefinitely. At 30% solids, sludge acts as a liquid; unstable and extremely low in strength. Approximately 25 million cubic metres of sludge at 30% solids are produced each year by the two operating extract ion plants owned by Syncrude Canada Ltd. and Suncor Inc. More than 500 million cubic metres of sludge have been produced over the first 20 years these plants have operated. The experiments detailed in this report show that it was possible to increase the solids content of sludge to 50% solids by adding three parts sand (tailings sand) to one part sludge. At 505 solids, the sand-sludge mixture was semi-plastic, but extremely weak. One thousand parts per million of lime were needed to keep the sand from segregating from the sludge. Drainage of sand-sludge mixtures, even under the pressure of self-consolidation, was slow and uneconomical. The sand-sludge mixture had to be dewatered to 85% solids content before its shear strength was sufficiently high to support machine traffic or the overboarding of more sand-sludge mixture. At 85% solids, the sand-sludge mixture had a shear strength in excess of 100 kPa. Freezing and thawing sludge (without sand) caused the solids content to increase from 30% to 50%. Another 10% increase in solids content was achieved by several more cycles of freezing and thawing. At 50% solids, sludge was semi-plastic. Ditches or grooves ploughed into the sludge remained, but the shear strength was very low (less than 2 kPa). Sludge without sand needed at least 80% solids to have sufficient shear strength (more than 100 kPa) to support machinery traffic or sludge overboarding. If snow was removed from the surface periodically, the sludge froze to 165 cm depth in one winter in Mildred Lake, the Syncrude Canada Ltd. plant and mine site, approximately 40 km north of fort McMurray, Alberta. If the snow cover was left in place, freezing was restricted to 30cm. Laboratory and pilot-plant experiments showed that the amount of sludge that could be frozen in one winter could be increased by freezing the sludge in thin layers. Using this technique, a layer only a few centimetres deep was deposited and left to freeze for a day or two; as soon as it was frozen, a second layer was deposited. Layered freezing was also slightly more effective at dewatering sludge than freezing a pool of sludge from the top down. The water released from the sludge during the thaw period rose almost immediately to the sludge surface. Surface water had to be drained away to allow further dewatering, either by evaporation or vegetation-controlled evapotranspiration. Standing water on the sludge surface prevented the establishment and growth of adapted vegetation by floating seeds, making the rooting medium unstable, or inhibiting oxygen flux to the root zone. If the water was removed, two species of plants—reed canary grass and western dock—were well adapted to the sludge environment and capable of removing enough water from the sludge to dry it to 80% solids. Reed canary grass was the best adapted plant to both sludge and sand-sludge mixtures. Furthermore, reed canary grass grew from small sections of its own rhizome, known as sprigs. Starting plants on sludge with sprigs of reed canary grass may allow for large scale (hundreds of hectares) dewatering by vegetation. Sprigs were easy to spread, not subject to movement by wind or small amounts of water, and fast to establish new plants. Sludge at 50% solids that was planted to reed canary grass was dewatered to 80% solids in one growing season. At 80% solids the sludge had a shear strength of 120 kPa and could support machine traffic of any kind or the overboarding of several metres of liquid sludge. However, the rapid removal of surface water and the quick establishment of a dense plant community were essential. Otherwise, dewatering during the summer months was minimal, less than a 5% increase in solids from May to October. Sand-sludge mixtures were also dewatered by freezing and thawing. A 1 year dewatering cycle that included freezing and thawing and summer evaporation, but no plant controlled evapotranspiration, increased the solids content of a 2-m deep sand-sludge mixture from 50% to 80% solids. Reed canary grass and western dock also grew well on sand-sludge mixtures and aided in dewatering, if the surface water was removed.