Sand will be considered only in connection with cement (Portland cement). See notes under caption, "Cement." The province of sand with cement is similar to that of brick in masonry, on a small scale. The cement acts in the same capacity as the mortar surrounding the brick. If a brick could be made the size of the complete wall there would be no need for mortar, but as it cannot, mortar is required to fill thoroughly all spaces between the bricks and to adhere to each tenaciously so as to form one solid mass. If on the sides of the brick should be placed thin layers of soft clay, the bond would be broken, inasmuch as the clay would not cling tenaciously to the brick and would prevent contact with the mortar; therefore, to obtain a good brick wall, each brick must be clean and free from coating of foreign material, so that the contact with the mortar will be complete.

To return to sand - in the same way, each particle must be entirely separate from any other, and must be free from foreign coatings which would prevent its contact with the cement or would cause the particles to cling together so as to prevent the thorough coating of each particle with cement. The usual requirement for sand to be used with cement is that it shall be clean, coarse, and sharp; and that rarely should over three volumes of sand be used to one volume of cement. In our case, however, we must be more economical - we find in our pits a soft, fine sand, which, while it separates well, resembles ground clay when rubbed between the fingers. Before throwing it out as useless and paying a round price for sand hauled in, let us see what can be done with it.

Make frames of wood screwed together which will provide 32 moulds for cement bricks 2 inches long, with 7/8-inch section, by using two long strips 7/8-inch square divided by 7/8-inch divisions so as to leave 2 inches clear between each, all to be oiled. In these spaces we are to mould our small bricks for test. (See Fig. 1.) Obtain some clean, coarse, sharp sand for comparison, and also some of the fine soft sand from the site; carefully mix (dry) enough cement with sand to form four bricks of each kind, first in proportions of 3 sand to 1 cement, then of 4 sand to 1 cement, then of 5 sand to 1 cement, and lastly of 6 sand to 1 cement - eight batches in all, one-half of which will be of sand from the site, the other of imported sand. Each batch should be mixed with greatest thoroughness in order that every side of every particle of sand shall become coated; then add just enough water to thoroughly dampen, and crowd the mixtures into the moulds (which should be laid on a smooth plank, for bottom), compressing each as much as possible. If water appears on top, it will indicate that too much has been used in mixing. Then there will be four bricks of each batch. Place on each an identifying mark.

Fig. 1. Frame for Moulding Cement Test Bricks.

Cover with a damp cloth to prevent evaporation; and after one day, draw out the screws holding the forms, and put the bricks for two weeks in a damp place, covered with a wet cloth or in water. Obtain two good, smooth planks about 3 feet long by 1 foot wide; go to a neighboring foundry, and obtain permission to use their scales and a few hundred pounds of pig iron; lay down one plank on a perfectly level bed, and place the number 8 brick on their sides. (See Fig. 2.)

Fig. 2. Method of Testing Cement Bricks for Crushing Strength.

In order to get an even bearing it is well to place below and over the cement bricks one layer of soft building paper; above place the second plank with great care, seeing that it has a firm, even bearing at each of the four bricks. Carefully lay pig iron above, beginning at the middle, and fill each way, never loading one end before or with a much heavier load than the other. Carefully watch the bricks; note when they crack; and when they crumble, weigh the pig iron and note the results.

As above, test each of the remaining seven sets of bricks.

It is safe for ordinary dwellings to select the cheapest materials and mixture which will carry under such conditions 500 pounds of iron.

For example, it is supposed that No. 3 sample, made up of 5 parts coarse sand to 1 part cement, carries the same load as No. 6 (4 parts sand from the site to 1 part cement). If it costs $1.50 for one cubic yard of coarse sand, and 50 cents a yard to haul the sand away from the excavation, the coarse sand costs us for use $2.00 a cubic yard, or 7.4 cents a cubic foot. With cement at $2.00 a barrel, we have:

Using sand from the site in mixture No. 6 (4 sand to 1 cement), we have the cost per cubic foot as follows:

It will thus be seen that it is cheaper to use the 4 to 1 mixture and the poorer sand from the site, than to haul it away, bring in other sand, and use less cement. If, on the other hand, it is found that it will require No. 5 to equal No. 3, it will be cheaper to haul away the excavated sand and haul in good sand.

Before breaking the bricks, they should be very carefully examined as to texture and firmness. Scratching with the finger-nail will show the general quality to the student after a short experience, so that experiment with weights is unnecessary after the student has become familiar with the appearance of good mortar.

After the bricks are broken, examine the fractures, and observe with a glass how well the spaces between the particles of sand are filled. Generally the results will be about in this order:

No. 1 will show much greater strength than the corresponding No. 5. No. 2 and No. 6 will show less difference. Possibly No. 3 will break under less weight than No. 7; while No. 4 will go to pieces much before No. 8. The reason for this is readily detected under a glass. The imported coarse sand requires much more cement to fill the spaces. In No. 1 they are all filled; while in No. 4 there are large voids, which do not occur to such an extent in Nos. 7 and 8; the sand being finer allows smaller voids. Observe that the increase in the number of voids or small holes in the mass weakens the mixture.

*Note. - When sand and cement are mixed in proper proportions, the bulk of the mixture is not greater than the hulk of the sand, as the cement only fills the voids or spaces between the particles of sand, just as there can be poured into a full measure of shot 20 per cent of water without overflowing the measure.

Now try one other experiment. Get two forms from which 4-inch cubes can be made; make a mixture of possibly 4 fine sand to 1 cement; mix one with a large amount of water, as is usually seen in ordinary work; and from this form a 4-inch cube. Mix for the other cube the same proportion, but with just enough water to change throughout the color of the mass. Crowd the mixtures into the form, both cubes to be compressed as much as possible; set the cubes aside, covered with a wet cloth for a week; then split with a cold chisel, and observe. The interior make-up of that mixed with a larger amount of water, is full of voids; that mixed with little water is compact.

The question of sand is treated, as later the question of roofs will be, in a great deal of home-made detail, not only to get at the question of sand and roofs, but to illustrate how, with materials that are always at hand, results of great practical value may be obtained.