In many forestry problems additive regression equations are expected because of the nature of the data. This paper discusses the reasons why one does or does not have these additive equations. Mathematical proofs of the theory are given, with a numerical example.