Superposition of signals in DNA molecule is a sufficiently general principle of information coding. The necessary re-quirement for such superposition is the degeneracy of the code, which allows placing different messages on the same DNA fragment. Code words that are equivalent in the informational sense (i.e., synonyms) form synonymous group and the entire set of code words is partitioned into synonymous groups. This paper is dedicated to constructing and analyzing the model of synonymous coding. We evaluate some characteristics of synonymous coding as applied to code words of length two although many definitions may be extended for words of arbitrary length.