Rotary kilns have been in service in the gold industry to regenerate carbon for 65 years. At the time oxide ores with gold head grades of 8 to 10 g/t were being treated. Carbon could be regenerated adequately at 650C and solution losses were less than 1%. Forty years ago, research indicated that heavily fouled carbon from flotation circuits had to be held at 750 to 850C for ten to twenty minutes for full reactivation (Avraamides and La Brooy, 1987), gold head grades were typically 5 g/t and solution losses typically 0.02 ppm - still less than 1%. In 2017 it was determined (Claflin et al., 2017) that a residence time of less than 1 minute at 750C can be adequate to reactivate carbon fouled with organics sufficiently to achieve 0.005 ppm solution losses in a flotation tailings leach plant.With the potential for discovery head grades to drop below 0.6 g/t, solution losses of 0.02 ppm represent over 3% loss. Tight operation of the whole carbon circuit is required to keep gold solution losses below 1% going forward.There are alternatives to a rotary carbon kiln, e.g. vertical kilns of which the Minfurn, Rintoul and Combustion Air kilns are examples. The alternative kilns can be heated using electrical heating elements, induction, and microwave. These kilns can operate successfully under the right conditions, however, they are almost exclusively for small carbon throughputs and are not suitable for the large carbon advance rates that are required to process low head grade ore deposits. Large throughput carbon kilns are are almost exclusively fossil fuel fired rotary kilns and the subject of this paper. In the view of the principal author, a fossil fuel fired vertical carbon kiln is preferred to a rotary kiln, however, it is not currently used in the industry.Thermal regeneration of activated carbon has a major impact on gold recovery in the carbon circuit. The carbon kiln and associated equipment constitute 3 to 5% of the capital cost for an oxide processing plant and up to 10% of the capital cost if flotation tails are leached. Despite the capital cost and impact on recovery, the carbon kiln is often poorly selected, operated and maintained.Six years ago Ausenco embarked on the Carbon Kiln Initiative to reduce solution gold losses. This body of knowledge has been published in a series of papers culminating in this paper on how to:1. Select a kiln appropriate for the flowsheet;2. Maintain the kiln to achieve design operation; and3. Operate the kiln to minimise costs and maximise reactivation.CITATION:Claflin, J K, La Brooy, S R and Claflin, A M, 2018. Carbon kiln selection, operation and maintenance, in Proceedings 14th AusIMM Mill Operators' Conference 2018, pp 251-274 (The Australasian Institute of Mining and Metallurgy: Melbourne).