This article attempts to prove the hypothesis on the “processual homogeneity” of the changes iR (*ŕ̥ *ir *yr) ≥ eR i iL ≥ eL. The source material for the following analysis has been excerpted from letters written in Polish between 1525–1550. Within the specified time brackets, the most dynamic is the process of the extension of the groups iR ≤ *ŕ̥, whereas the extended combinations eR ≤ *ir *yr are characterized by far lower intensity. The extensions within the groups of the type eL are not numerous and rare. Textual extensions of the combinations eR and eL show a significant convergence: forms with eL are to be found mostly in those letters that consistently demonstrate and provide evidence of the occurrence of the group eR; while they are virtually non-existent in the texts that document exclusively the combination iR. This fact confirms the hypothesis that combines both types of change into one process. The author does not associate the dependencies in the different pace of both phenomena and their different outcome with phonetic factors, but with the sensitivity to phonotactic and phonostatistical provisions and patterns, as well as with the morphological placement of the dyads under scrutiny. This interpretation, in turn, provides a new insight into the nature of the process of shaping linguistic norms: all the contributing processes have random (non-purposeful) origins, and their constraints involve a necessity to retain the necessary minimum of the internal balance.