40 BULLETIX 463, U. S. DEPARTMENT OF AGRICULTURE.. 
determined for each sample prior to the time of making the slaking 
test, so that the percentage may be corrected by adding sand or clay. 
This test may be employed also to compare the slaking qualities of 
the same clay when mixed with different percentages of sand, and 
when so employed may afford a valuable guide in fixing the proper 
proportions for an artificial sand-clay mixture. In making the test 
several balls should be made and tested from each kind of clay, so 
as to afford check results. Also, in all the tests care must be exercised 
to keep the different samples from becoming confused after the balls 
are placed in the water. To do this, identifying marks may be placed 
upon the balls, or the balls from each sample may be placed in a 
different part of the containing vessel. 
(3) The amounj; which clay shrinks on drying out also may be a 
measure of its fitness for use in a sand-clay surface. The shrinkage 
may be tested by observing the behavior of the clay balls while they 
are being dried out. when those which shrink most will develop most 
cracks. Other things being equal, the sand-clay mixture showing the 
smallest shrinkage is best suited for use in a road surface. 
(4) Valuable comparative information regarding different avail- 
able clays or sand-clay mixtures may be gained sometimes by simply 
observing exposed surfaces, where the materials outcrop, under 
different weather conditions. The best clays will be capable of stand- 
ing on relatively steep slopes and will develop fewer surface cracks 
upon drying out after a wet season. They also will be little affected 
by frost and will appear dense and firm in dry weather. 
(5) To test the suitability of sand for use in a sand-clay surface, 
place a sample of the sand in a vessel containing water and agitate 
the water until the sand is thoroughly in suspension. Then, after 
the sand has been allowed a few moments to settle, pour off the water 
slowly. If of good quality, the sand will not be carried out with 
the water but will remain in the vessel until practically all of the 
water has been drained off. Sand containing a large percentage of 
mica or other light mineral matter will not meet this test and is not 
generally suitable for use. 
(6) The methods described above for testing sand-clay materials 
are based on the assumption that the selection must be restricted to 
materials locally available and are intended simply to aid in com- 
paring the relative merits of such materials. But these methods 
can. of course, be adapted to more exact laboratory practice if so 
desired. 1 
1 A rather detailed plan for laboratory procedure, which has been followed to some 
extent in connection with the Band-clay roads of Georgia, is described by John C. Koch 
in the " Transactions of the American Society of Civil Engineers. " Vol. LXXVII, Decem- 
ber, 1914, p. 1454. 
