cJb BULLETIN 704, U. S. DEPAETMEXT OF AGRICULTUEE. 



from the bins themselves by means of proper sampling lubes, provided samples 

 are taken from points well distributed over the face of the bins ; or from the 

 bins at the point of discharge, provided sufficient material is drawn from each 

 discharge to obtain samples representative of the material in the bins. In no 

 case should a single composite sample represent over 200 barrels. Samples 

 should be shipped and stored in air-tight containers. Before testing, they 

 should be passed through a 20-mesh sieve in order to thoroughly mix the 

 sample, break up lumps, and remove foreign material. 



PAVING BRICK. 



Samples of vitrified paving brick may be taken at the point of manufacture 

 or from cars at the point of delivery. Samples from the plant should preferably 

 be taken from the kiln at the time of emptying. One or more sets of tests, de- 

 pending upon the size of the kiln, each set consisting of three separate tests, 

 should be made on each kiln. Each test in a set of three should represent 

 approximately a single degree of bux'ning (based on the position of the brick 

 in the kiln), and all 10 of the brick in a single test should be of the same 

 approximate degi*ee of burning. In general, samples selected from piles at the 

 plant should be as nearly as possible representative of the entire run of the 

 brick. Samples from piles should be taken from as many different points corre- 

 sponding to the length, breadth, and depth of the pile as possible. In no case 

 should they be confined to the upper or outer few laj'^ers. Where controversy 

 arises regarding the admissibility of certain types or portions of the lot, en- 

 tire test samples may be selected from such types or portions having a char- 

 acteristic appearance in common. When sampled at the point of delivery, a 

 representative sample should be taken from each carload received. Consid- 

 erations covered under sampling from piles at the plant apply equally to sam- 

 pling from cars. No bricks should be included in the test lot which would be 

 rejected on the basis of cracks, chips, or other defects covered by the specifica- 

 tion clauses for visual inspection. Samples should be shipped in stout boxes or 

 crates in lots of 12, put up in 2 rows of 6 bricks each, separated by a wood 

 partition. 



STONE BLOCK. 



Stone block may be sampled for quality and size either at the quarry or from 

 cars. A preliminary sample for quality consisting of at least four standard 

 blocks should be submitted from each proposed source of supply at least two 

 weeks prior to the date of acceptance or rejection. Additional samples of 

 blocks may be taken from time to time during the progress of the work, when- 

 ever the quality or appearance of the blocks varies, and at such other times as 

 may be directed. No sample should include blocks which would be rejected on 

 visual inspection. Preliminary samples may be omitted in cases where material 

 from the proposed source of supply has been tested within one year prior to the 

 date of acceptance or rejection, in which cases the report of such tests may be 

 use<l as the basis for acceptance or rejection. Samples of stone blocks should 

 be shipped in stout boxes or crates. 



