placed in the vessel from dropping down into the pipe. The pipe b is connected by a rubber hose to a 5-gallon water bottle. The lid c is of cast iron and is ground to an air-tight fit. The small cock d is connected by a rubber tube to a Richards air pump (shown at d, Pl. IV, B, and at b, Pl. VI, A), which is attached to a water faucet. Pl. IV, B, shows the apparatus for filling the cylinder, with the water bottle, the air pump d, and the scales in place. The material is placed in the hopper a and runs through the opening at b into the cylinder c.

Method. A sample of the material is dried to constant weight, thus incidentally obtaining the percentage of moisture contained in the material as received. The dry material is placed in the hopper (a, Pl. IV, B), which is raised so that its bottom is 24 inches above the center of the cylinder. The aperture b is opened to such an extent that the material will fill the cylinder in about one minute. The material is allowed to fill the cylinder to overflowing and is struck off level with a straightedge. The cylinder is then transferred to the scales by means of the overhead trolley (Pl. IV, A) and its weight with contents is ascertained and recorded. The tube from the water bottle is next attached to the pipe beneath the cylinder and the tube from the pump is attached to the valve in the lid, which is carefully put in place. The air is exhausted from the cylinder by means of the air pump (d, Pl. IV, B), and the water is allowed to flow in slowly from below until it approaches the top, when the cover is removed and the water brought to the top of the vessel. The water is allowed to flow into the voids at such a rate that it reaches the top in about one hour. The weight of the vessel is taken after the voids are filled with water.

The percentage of voids in the large material and in the screenings and sand is determined in the same way except that a smaller cylinder is used for the material sifted to one size. This cylinder has only one hole in the bottom for the introduction of the water. The smaller cylinder is used because the material of a given size is often insufficient to fill the larger one (shown in Pl. V, A). Pl. V, B, shows the test on large-sized material under way.

Computations. The volume and weight of the vessel and the weight. of water required to fill the pipes beneath the vessel are known. The quantities measured and recorded during the tests are (1) the weight of the vessel full of dry material, (2) the weight of the vessel full of material and water, (3) the weight of a portion of the surfacedried sand after the test, and (4) the weight of the thoroughly dried sand after the test.

The percentage of voids is obtained by correcting the measured volume of water for that filling the pipes and absorbed by the material, dividing by the volume of the vessel and multiplying the result by 100.

The percentage of voids is also obtained by dividing the weight per cubic foot of the dried sample by the product of the specific gravity of this material and the weight of a cubic foot of water.

PERCENTAGE OF MOISTURE.

Method. The weight of water introduced into the cylinder is corrected for the absorption of water by the stone, which is determined by weighing a portion of the material after surface drying and again after thoroughly drying it over a hot plate.

Computations. The percentage of moisture is found by dividing the difference in weight between an ordinary sample before and after drying by the weight of the dry material and multiplying by 100.

WEIGHT PER CUBIC FOOT.

The weight per cubic foot of the material is found by taking the difference between the weight of the empty vessel and the weight of the vessel full of dry material and dividing by the volume of the vessel.

PERCENTAGE OF ABSORPTION.

Method. The sample used for the absorption test, taken from the material in the void apparatus, is obtained by spreading the material out and selecting small portions from several different parts of the

mass.

Computations. The unit absorption, or the weight in pounds of water absorbed per pound of dry material, is the difference between the weight of the surface-dried material after the test and of the thoroughly dried material divided by the weight of the thoroughly dried material.

SPECIFIC GRAVITY.

METHOD USED WITH LARGE MATERIAL.

Apparatus.—Pl. VI, C, shows the apparatus used in determining the specific gravity of large material. The cylinder (a) is the same as that used for making the void test on "one-size" material. It is made from an 8-inch wrought-iron pipe 8 inches long; the cap at the bottom has a 1-inch opening in the center fitted with pipe and valve (shown at b) and is protected from the materials placed in the cylinder by a No. 50 wire-gauze screen. This pipe is connected by rubber tubing to an aspirator bottle containing clean water. The valve in the lid is connected to the Richards air pump (shown at c). The lid is of cast iron and is grooved to fit the top of the cylinder. A rubber gasket is placed in the groove, making an air-tight joint. The tin vessel (d) is arranged to hang from the arm of the balance (e), so that its weight may be determined when it is immersed in water.

B. APPARATUS FOR REMOVING
SILT FROM STONE SCREEN-
INGS.

(See explanation in text, p. 28.)

(See explanation in text, p. 26.)