Abstract-In spite of the former remarks, there are relatively few contributions in the literature related to real multiplevalued reversible circuits, as compared to contributions on binary reversible circuits. Most, if not all of them, are dedicated to ternary reversible circuits [3,15,16,20]. Some careful basic mathematical studies support these efforts (see e.g. [1,4,10,11]).Reversible circuits are realized with fan-out free, feedback free, cascades of reversible gates, where reversible gates realize bijections on a given set of p n p-valued n-tuples. These constraints suggest that the design of reversible circuits is much more difficult than that of classical ("irreversible") digital circuits. This is one important reason why evolutionary approaches for the design of binary and MV reversible circuits are receiving increasing attention (see e.g. [5,6,13,14]).The present paper has been mainly motivated by the pair of nicely selfcontained contributions [15,16], (where basic gates, a circuits design algorithm, a non-trivial test circuit, and details of low level realization are discussed), and by some open questions and suggestions found in an unfortunately unfinished-unpublished paper of Marek Perkowski and colleagues [20].The next section reviews some mathematical aspects, Pauli matrices, representation of the basic ternary values in the Bloch sphere, and the role of the VilenkinChrestenson transform. In a following section an analysis is done to determine whether Barenco et al. type of structures [2], and Sasanian-Wang-Perkowski type of structures [22,24] may be used in ternary reversible computing, with similar advantages as in the binary case. A last section is devoted to introduce the Entanglement phenomenon and its presence in ternary quantum computing. Some conclusions will close the paper.