In a previous work [1] we emphasised about the necessity of shielding the potassium trisoxalatoferrate crystals during the reactor irradiation. Indeed owing the gamma flux associated to the neutron flux in reactor, the substance to be irradiated may recive a great gamma dose and so to be in some extent radiolytically decomposed. The lead shield we used had a thickness of 2.5-3 cm; but owing the long irradiation time and great sensitivity of potassium trisoxalatoferrate, there was the danger of some radiolytic decomposition. In connection with this problem it should be interesting to know in what extent the activity of ferrous species [2] is due to the recoil or to the radiolysis process. A shure answer to this question may be given only after an especial study on gamma radiolysis of the potassium trisoxalatoferrate crystals. In a recent paper SIEKIERSKA and FENGER [3] found that after 24 h of maintaining the potassium trisoxalatoferrate crystals (irradiated in carbonic snow) to the room temperature, the Fe 2 + is already in a great extent oxidised to trivalent state. They thought that the différencies in the irradiation conditions explain the différencies of the temperature range in the oxidation of ferrous species, obtained in [2] and [3]. This problem needs a more thoroughly discussion. The separation method used by SIEKIERSKA and FEN-GER [4] allows to separate only the Fe 2 + in the sum of Fe 3 + plus Fe(III) complex. In our opinion [5] it is not possible to give a correct kinetic interpretation without a separation of the Fe 3 + and Fe(III) complex species. Is the reason for why we used an ion-exchange method, allowing the separation of three species: Fe 2 +, Fe 3+ and Fe(III) complex. But even in the plotting manner used by SIEKIERSKA and FENGER [3] the results of the experiences we carried out recently by irradiation to Ris0-Reactor in liquid nitrogen show that the oxidation of Fe 2+ to Fe 3+ is not a fast process and the Fe 2+ quantity is time dependent (Fig. 1 ). In the case of curves 1 and 2 the crystals were irradiated in liquid nitrogen and the curve 3 is represented after SIEKIERSKA and FENGER'S results (irradiation in carbonic snow) [3]. It is possible to remark a considerable effect of time heating before the chemical analysis: the plateau for 1 h is located at 21%, while for 24 h the plateau is located at 4.5% Fe 2+ . So it is not sufficient to observe that the Fe 2+ is oxidised, without to take into consideration the time heating before the chemical analysis. The value at -78 °C (Fig. I) was obtained by dissolving the crystals in water at 2-4 °C [3]. An extrapolation of the results over 0 °C to this value may be not maded because the dissolution at a such great temperature modify in an uncontrolable manner the change of Fe 2 + quantity. So, for a sample irradiated at -200 °C and disolved at 8-12 °C we obtained only 41 % Fe 2 +, a very small value comparated with 65% in the previous case. Inaccesible temperature domain for chemical analysis in aqueous medium -200°C -o -79°C a g g «'[%] S Inaccesibl...