20 yen = § : \ re 3 20F SA 5 = , 2 s 6 se DF 1Scmisec 7 ~ 29 = 10%%/em? ~w, (equiv. ae ete 10F Ra a8 2 6) Toe Uses of Atomic Energy) (United Nations, Geneva, 1958), vol. 18, p. 351. and B. Szabe, J. Oceanogr. Soc. Japan, 20th anniversary volume (1962): H. Pettersson, Papers in Marine Biology and Oceanography [suppl. to Deep-Sea Res. 3 (1955)], p. 335; Y. Miyake and Y. Sigimura, Studies on Oceanography (Editorial Comm. for Sugawara festival vol., Nagoya, Japan, 1964), p. 274. 4. W. S. Broecker, R. Gerard, W. H. Ewing, B. C. Heezen, J, Geophys. Res. 65, 2903 (1960), 5. T. J. Chow and E. D. Goldberg, Geochim. Cosmochim. Acta 20, 192 (1960). agi, QO 1 10 1 20 1 30 au 6. G. 5. Bien, N. W. Rakestraw, H. E. Suess, Excess Rn? (roequiv. gram R226) o@-~l . Oo Fig. 3. Excess Rn22? as a function of distance above the sea bottom in the South Atlantic at 22°47'S, 32°37'W. . . 10. In our previous report it was shown that the following relation existed be- tween the radon excess, C,*, and the distance from the bottom. rn C,* =M Bs — exp (—x VA/DzE) where M is the standing crop of radon lost by the sediments, and » and Dz are as defined above. Data for the profile 23°S, 33°W in the South Atlantic are plotted in Fig. 3. The points can be fit by the theoretical equation if Dg is 1.5 cm?/sec and M is 9 X 10-3 gram equivalent of Ra?** per liter. As previously published (/) measurements of radon leakage from the tops of triggerweight cores yield standing crops of from 1.5 to 35 & 10-13 gram equiva- lent of Ra®?* per square centimeter. This amount of radon is also what would be expected from molecular diffusion from the radon-rich pore water [see (7)]. For the other two profiles roughly the same standing crop of excess radon is required, but the eddy diffusion rate must be about 30 times higher (that is, ~50 cm?/sec). Hence the study of the distribution of excess radon near the ocean bottom will prove to be a powerful tool in determining rates of vertical mixing. WALLACE S. BROECKER Yuan Hui Li JOHN CROMWELL Lamont Geological Observatory, Columbia University, Palisades, New York References and Notes Bull. Inst. GQeeanogr. Monaco 61, No. 1278 (1963); Nuclear Geophysics (Nat. Acad. Sci.Nat. Res. Counc. publ. 1075, 1963), p. 152. W. H. Munk, Deep-Sea Res. 13, 707 (1966). D. Lal, E. D. Goldberg, M. Koide, Science 131, 332 (1960). R. Gerard and M. Ewing, Deep-Sea Res. 8, 298 (1961). Supported by Atomic Energy Commission contracts AT(30-1)2493 and AT(30-1)2663. We appreciate the help of F. Gwinner and G. Parker in constructing and maintaining the apparatus. Only because of the help and cooperation of many members of the staff conNected with the operation of the Lamont research vessels Verna and Conrad could these Measurements have been carried out. Lamont Geological Observatory contribution No. 1129. 23 August 1967 a Silica in Alkaline Brines Abstract. Analysis of sodium car- bonate-bicarbonate brines from closed basins in volcanic terranes of Oregon and Kenya reveals silica contents of up to 2700 parts per million at pH’s higher than 10, These high concentra- tions of SiO, can be attributed to reaction of waters with silicates, and subsequent evaporative concentration accompanied by a rise in pH. SuperSaturation with respect to amorphous silica may occur and persist for brines that are out of contact with silicate muds and undersaturated with respect to trona; correlation of SiO, with concentration of Na and total CO, support this interpretation. Addition of moredilute waters to alkaline brines may lower the pH and cause inorganic pre- cipitation of substantial amounts of silica. The SiO, content of natural waters that are not associated with areas of geothermal activity very rarely exceeds 100 parts per million (ppm) (7). Com- monly accepted saturation values with respect to amorphoussilica at 25°C and Silica contents of brines from four closed basins, three of which belong to the Great Basin province of the western United States, have been plotted against pH (Fig. 1), Na (Fig. 2), and total CO, when sufficient sample was available (Fig. 3). Silica was determined colorimetrically as the B-silicomolybdate complex (4). We have paid special attention to pH control and elimination of reductants, and have used tartaric acid to eliminate phosphate interference (5). All analyses were repeated several times and checked against synthetic brines of similar composition. Interfer- ence by boron or fluoride was negligible. Exceptionally high values of SiO, were checked by gravimetric and fusion procedures. Brines were stored in polyethylene bottles and filtered under pressure through 0.45-~. membranes prior to analysis, although filtration through pore sizes down to less than 0.1 p showed noeffect on silica content. Pre- cipitation of silica during storage was -checked by digestion of the complete bottles of duplicate samples. Differences were usually less than 5 percent of the silica content at high concentrations of SiO., and much less at low concentrations. The western Great Basin has several intermontane areas of interior drainage in igneous rock terranes. Most of these basins contain saline lakes or playas with brines high in carbonate; the brines have been derived principally by evaporative concentration. Sodium and carbonate are the dominantions in solution because alkaline-earth carbonates pre- cipitate, and sources of sulfate or chloride are lacking. Abert Lake of south-central Oregon, with an area of about 130 km?, is one of several highly saline remnants of the large pluvial lakes that once occupied the western Great Basin (6). The present lake has no outlet, and its level and total surface area reflect the long-term balance between inflow, principally the Chewaucan River, and loss of water by evaporation. Sodium, carbonate species, and chloride ions comprise over 90 percent of the dissolved constituents in precipitation of silica (3). In recent addition to waters from Abert Lake countered waters having exceptionally sociated lacustrine sediments, we have included a sample from a brine pond on the salt flats at the north end of the thought to have formed by inorganic studies of alkali carbonate brines of a number of closed basins we have en- 2. F. F. Koczy, in Natural Radium as a Tracer in’ the Ocean (U.N. Int. Conf. Peaceful high silica contents—up to 2700 ppm. 1310 for the inorganic precipitation of chert. a pH lower than 9.2 are 110 to 140 ppm (2). Some chert deposits are 1. W. S. Broecker, in Symposium on Diffusion in Oceans and Fresh Waters, T. Ichite, Ed, (Lamont Geological Observatory, Palisades, N.Y., 1964), pp. 116-145, These data suggest a simple mechanism Abert Lake and associated springs. In itself, and interstitial solutions from as- 4 SCIENCE, VOL. 158