In this paper the thermal stability of poly (A + U) and poly (A + 2U) in acid solution is studied. It is known that at low pH poly A associates with itself to form a double protonated hellcat structure, poly (A + A). The dissociation Tm's of poly (A + U) and poly (A + 2U) do not vary with pH as long as they are higher than TmA [the dissociation temperature of poly (A + A)]. TmA itself varies with the pH (it increases as the pit is lowered) and thus, there are pH values at which TmA equals the dissociation Tm'S of the complexes poly (A + U) and poly (A + 2U). Below such pH's, Tm2-3, Tm3-1, Tm2-1 decrease as the pit becomes more acid. In contrast, Tm3-2 does not exhibit such a variation. The transition corresponding to this last Tm [the partial dissociation of poly (A + 2U) into poly (A + U)] does not change the concentration of ‘free’ poly A. We conclude that the change in thermal stability of the complexes in this region of pH is due only to their competition with poly (A + A) and that poly (A + U) and poly (A + 2U) are not modified. At low ionic strength, the transitions observed change with the pH and, by identifying them by means of two independent methods (continuous variation method, and comparison of spectral variations) we have shown that the stability diagram (temperature versus pH) formally resembles the stability diagram (temperature versus ionic strength) at neutral pH (Fig.3), when pH is decreased and/or ionic strength is increased, poly (A + 2U) is relatively more stable than poly (A + U): The formation of poly (A + A) lowers the stability of both poly (A + U) and poly (A + 2U), but, since the proportion of adenylic residues is 1/3 in this last complex, against 1/2 in the former, poly (A + 2U) is less destabilized, and is more sensitive to the diminution of electrostatic repulsion brought about by the sodium ion concentration. These effects are summarized in a three-dimensional diagram temperature-pH-ionic strength (Fig. 9). Some of the transitions occuring in acidic solution appear to be irreversible, or very slowly reversible. Such is the case of the formation of poly (A + U). We have shown that this effect is due to the structure poly (A + A). When the dissociation temperature (TmA) of this structure is higher than that of poly (A + U), it is necessary to Porto two A — U pairs replacing one A — A pair in order to obtain the free energy decrease. Therefore, such a reaction cannot occur in two steps, and this suggests a third order mechanism. Another example is given by the rearrangement of poly (A + U) into poly (A + 2U) and poly (A + A). This observation leads us to a general rule concerning the possibility of reorganization in a system of polynucleotides where several associations are possible [for instance it answers the question as to why poly G cannot expel poly I from poly (I + C)]. [ABSTRACT FROM AUTHOR]