1454 PERSISTENCE OF TRITIUM AND #C IN THE PACIFIC PROVING GROUND recent studies at the Chalk River Laboratories of the Atomic Energy of Canada Limited,@” concerned with dating tritium dispersal in ground waterby tree-ring analysis, indicate that part of the tissue-bound tritium of cellulose is held as exchangeable hydroxyl ions. The effect of these two processes on the tissue-bound tritium data reported in this paper has not been evaluated at this time. The state of tritium in soil and geological materials, especially those modificd and dis- placed by a nuclear detonation, has not been adequately described. According to RANKAMA and Sanama,'") hydrogen is found in mineral structures in the four following states: independent hydroxyls, structural water, hygroscopic water and hydrides. Rock and soil materials also contain interstitial water which is held by capillary forces. Structural water and hygro- scopic water, in addition to interstitial or capillary water, are the most important to an understanding of the problem considered. The water usually most available to plants from the soil is capillary and hygroscopic in form. Hygroscopic water is generally assumed to be available to plants to tensions of 31 atm (hygroscopic coefficient) and capillary wateris readily available for absorption by roots. Structural water cannot be removed without breaking down the structure of the mineral, but there is biological evidence in this paper and elsewhere that structural water exchanges at some unknown rate with the free capillary and hygroscopic water of the soil or mineral materials. The movement of tritiated water through geological strata has been studied at the Idaho Reactor Test site in connection with the monitoring of waste disposal wells.2) An attenuation of tritium concentrations in the ground water was partially attributed to the exchange of tritiated water, with water of hydration or similarly bound water in the rock strata. Ruyopes and Wirprnc*) conducted experiments with exchange columnsof various rock types and alluvial sediments, and passed tritiated water through them. Even with small columns, an attenuation oftritium concentration occurred as the solution passed through the columns. The Eniwetok soil materials are composed almost entirely of coral fragments and debris (CaCO, and Ca(OH),). The work of Apany, Fariow and ScHE."®) leads to the conclusion . that these two compoundsare the sourceofthe ‘ high residual tritium and *4C in the soil samplg from Eniwetok Atoll. Apams, FaRLow and SCHELLstudied large particles formed in nuclea, detonations at Eniwetok, Bikini, and the Nevaq, Test Site. Facts from the above data will help to inter. pret the preliminary data in Tables 1-9. They facts are: (a) loosely-held tritiated water present in th root zone of vascular plants enters and moig within the plant along conventional physiolog. ical pathways, (b) various tissues of the plant exhibit , dynamic near-equilibrium relationship with the tritium (THO)ofthe root zone, (c) changes in the tritium concentration « the available soil water will be reflected in, short time in the loose-water tritium content: someplant tissues, and (d) tritium may become bound in falla and substratum materials of the detonation s+ : in a tightly-held hydrated or chemically bourc state because of the fireball and post-detonaucr. physicochemical phenomena. Interpretation of results The reproducibility of the data reported .; at ¢ rele wat the con tran valu rium the ! tree is WI the § expe Tr the 3 the C is no moist sidere The samp! is tril. menta higher in peri to the tinuum af Mess at earl arc cc this paperis illustrated by the values in Table + content duplicated samples from the same specimen «- compar ereen le These data were obtained from the analvsy original sample. The water samples sh- excellent agreement and are both within tic- prescribed errors. The wood samples, in genera show good agreement except for the one sample which showsanerror by a factor of 2. Since. was not possible to obtain duplicate samples every case, the data must be interpreted wic this variation in mind. The data in Tables |, 2 and 3 indicate thai. ; most of the environments from which thesamp~ | were obtained, a considerable source of trius7 | exists in a boundstate within the soil materiae andis slowly exchanging with the loose waters the soil. As expected, an increased level « tritium in the loose soil water produced a increasedlevel oftritium in the loose waterofthe leaves of plants growing on thatsoil. Theloo water tritium ofMesserschmidia wood (3230 T.U varying 12-l5 n On E water in contentc available tritium | 4s much ‘urrent kc Becaus svatlable sources be the organ to be anc events she above. T, ieadily in vid rapid!