It is desirable to run off the upper third of the column at intervals of fifteen to twenty minutes by temporarily increasing the velocity. Recent sediments, river alluvium, etc., are more easily separated than soils of more ancient formation. The second, third, etc., separations are naturally accomplished in much less time than the first. The respective velocities of the separations should be 0.25 millimeter, 0.50 millimeter, one millimeter, two millimeters, four millimeters, eight millimeters, sixteen millimeters, thirty-two millimeters, and sixty-four millimeters a second. Below a velocity of four millimeters a second the mechanical stirrer is indispensable. Above this velocity the current of water in the conical base will be sufficient to bring the desired particles into the ascending column. At this velocity also a smaller elutriating tube having one-half or one-quarter the cross-section of the first may be employed to hasten the operation and diminish the quantity of water required. The quantity of water required for a complete separation is from 100 to 120 liters. Any soft water free of organic matter may be used, but distilled water is best. Hard water should be avoided. The mean time required for the different separations is as follows: 0.25 millimeter hydraulic value, thirty-five hours; 0.50 millimeter hydraulic value, twenty hours; one millimeter hydraulic value, seven and a half hours; two to sixty-four millimeters hydraulic value, eight hours. With proper arrangements for night work, an analysis may be finished in three or four days not counting the time required for the previous separation of the clay. =231. Weighing the Sediments.=—The sediments should be dried at the same temperature used for drying the soils. Hilgard dries both at 100°. Great care should be used in weighing the exceedingly hygroscopic clay sediments. In the case of the sediment of 0.25 millimeter hydraulic value it is allowed to subside as much as possible and after removing the supernatant water the residue, twenty-five to fifty cubic centimeters, is evaporated in a platinum dish and weighed therein. The water can be completely decanted from the other sediments, and they can be dried and weighed without any unusual precautions. The loss in the separation of clays and subsoils containing but little organic matter is usually from 1.5 to 2 per cent. This loss is partly due to the fine silt which comes off during the whole of the process and which is lost in the decanted waters of the sediments of 0.25 millimeters hydraulic value and above. The procedures indicated above are not strictly applicable to soils rich in humus and other organic matters, but the destruction of these matters by ignition leaves the residual soil in a condition wholly unfit for sedimentary separation.