This paper discusses the general implementation and choices for a SCHEME runtime structure on a RISC processor. It includes a general overview of the work presented, the register usage, the object types and their tags, the procedure calls, the arithmetic operations, and a quick description of the garbage collection. The authors present the practical aspect of their work and illustrate it with samples of machine code, including details of timing. In some cases, they compare this timing with that of a LISP compiler for an IBM RT. It would have been more interesting if this comparison had included a non-RISC processor. The authors include a list of the problematic features, which could be helpful for anyone starting a compiler project with this processor. This paper could serve as a reference for people who want to implement a functional language on this type of processor.