Three apparently conflicting generalizations concerning the relationships between molecules and cell structure may be derived from studies on cellular patterning in the ciliates. (A) Cells with identical genes and molecular composition may have different hereditary patterns. (B) Genes, through their prescribed molecular derivatives, constrain the modes of pattern permutation and define the states of greatest stability. (C) Cells with identical hereditary patterns may have entirely different genes and molecular compostitons. These priniciples may be reconciled through the recognition that they are characteristically applicable over different time intervals. Hereditary differences within a clone and without related molecular differences (principle A) may persist for hundreds of cell generations, but they are resolved eventually within a constant environment (principle B) as the configuration of minimal free energy is approached. On an even longer time scale, molecular substitutions have occurred for many or most components of the cell, but these have been constrained by selective pressures on an ancient design (principle C) that disallow substitutions affecting certain form-function relations which have been elevated to an adaptive peak.