Ten grams of North Haven brick clay were boiled continuously for nine days with about 700 cubic centimeters of distilled water, in a glass flask of one liter capacity and furnished with a reflux condenser. Fifteen grams were boiled in the same manner for eight and one-half days. When the boiling was concluded, the soil was found to have assumed a granular condition, the clay and fine sand being collected into a mass of small grains resembling coarse sand and settling rapidly. One portion thus boiled was elutriated by the beaker method, the other by Hilgard’s. The pestle was not used on either of those portions as it was desired to determine simply the effect of prolonged boiling. The separations thus accomplished are here compared with the elutriations of the same soil boiled twenty-three hours and of the pestled but unboiled soil. Hilgard elutriation. Beaker elutriation. Not pestled. Not pestled. Not Pestled. Pestled. Boiled Boiled eight twenty-three Boiled nine and a half hours, days, days. Not boiled. Diameter of particles. per cent. per cent. per cent. per cent. By sieves 3.36 3.24 3.63 3.49 0.25–0.05 1.21 1.11 1.91 1.29 0.05–0.01 28.27 33.04 33.61 27.02 0.01–0 56.29 48.85 54.78 52.21 Clay 4.92 3.05 1.97 10.15 Loss on ignition 5.95 5.95 5.95 5.95 —————— —————— —————— —————— 100.00 95.24 101.85 100.11 Here it is observed that the eight to nine days boiling diminished the clay as determined by Hilgard’s conventional method by seven to eight per cent, increasing the dust by two to three per cent, and the silt by about six per cent. Under the microscope small, rounded, opaque, brown granules were seen in large numbers, which when pressed under the cover glass, broke up into a multitude of very fine particles. From these experiments it would appear to be conclusively proved that too long boiling precipitates clay and thereby defeats the very object of the operation. In these experiments the time of boiling was prolonged in order to bring out unmistakably the effects of this operation. If ebullition for eight or nine days reduces clay from ten to two per cent, increasing the 0.05–0.01 millimeter diameter grades by six per cent, it is evident that boiling for one day or a shorter time becomes a questionable treatment. Further experiments[158] made by boiling clay in a platinum vessel with a platinum condenser showed that this precipitation of the clay was largely if not wholly due to the salts extracted from the soil.