SERVICE TKSTS. 15 



which the exact conditions of service may or may not l>c approximated, 

 according as it is diflicult or impracticable to duplicate them. The 

 essential requirement of such laboratory te>N is that they shall he 

 uniform, well defined, and directed to the measurement of the IIM-- 

 ful properties of the material. The samples must he submitted to 

 the conditions which are exactly known and the elements or combina- 

 tion of elements of endurance and value isolated and determined 

 quantitatively. 



The nature of the agencies of service being known, and the degree 

 to which the material possesses mechanical properties useful in 

 resisting these agencies having been determined by test, the value of 

 the material for any given service may be predicted. In case mate- 

 rials do not differ widely in some element of value it is necessary to 

 resort to known and constant conditions of laboratory tests in order 

 to determine their relative merit. The reliability of laboratory tests 

 may be determined by applying them to appropriate materials the 

 qualitv of which is known from experience in service. It is well to 

 remember that what are apparently small differences of quality in a 

 material will result in large differences of expense of maintenance and 

 of road life. It is not too much to say that the life of a macadam road 

 costing from $4,000 to $10,000 per mile may be from two to twenty 

 years, depending upon a proper choice of material to suit the local 

 conditions. The need and usefulness of proper laboratory tests to 

 determine the essential qualities of road materials is thus evident. 



Certain formal competitive tests of materials are, however, to be 

 noted. Various brands of paving brick have been laid in the same 

 street with a view to determining their relative merit and to check the 

 results of the rattler test. Such tests are in progress in the cities of 

 Washington, Baltimore, and Detroit. 



In France elaborate data have been collected. In 1865 tesN were 

 begun on the national roads of France to determine the quality of the 

 various materials of construction and ultimately to ascertain for each 

 department the materials best adapted for its roads. Careful meas- 

 urements were taken of the depth of stone for all the national roads 

 by digging trenches, alternately from one side of the road to the cen- 

 ter, so as to give a complete section of one-half of the road. This 

 test was repeated in 1871, 1886, and 1803, the last measurement- 

 requiring over a half million trenches. These measurements were 

 taken for the purpose of determining accurately the amount of -tone 

 worn off annually. At the same time a census was taken of the traffic 

 passing- over each road. Each division engineer was also required to 

 build short sections of road of each available rock in his district along 

 the same line of travel. All the materials of construction and main- 

 tenance were carefully measured and the results of wear observed. 



