326 Journal of Agricultural Research voi.x.no. 7 



however, they yield toughness results which exceed, in some cases, those 

 obtained with the solid rock itself. The limestone, with a toughness of 

 6, in only one case shows as low a toughness as this with a binder, and in 

 combination with the refined water-gas tar Y develops a toughness of 

 17. The chlorite gneiss (871), which has a toughness of 3, in no case 

 has as low a toughness as this when combined with any of the bitumi- 

 nous binders which were used; and in combination with the refined 

 water-gas tar Y and the fluxed native asphalt, it yields toughness results 

 as high as 17 and 15, respectively. A reference to Tables V and VIII 

 shows the same behavior with most of the binders for the biotite schist, 

 biotite gneiss, and, more noticeably, for the limestone. In comparing 

 these tables it will be noted that all the binders fail to produce a relatively 

 high toughness with the feldspathic sandstone, whose service results 

 were referred to above. From these results it would appear that some 

 modification of the toughness limits for bituminous macadam suggested 

 by Hubbard and Jackson ^ could well be made to admit rocks of lower 

 toughness than they would otherwise admit, provided such rocks can be 

 made to develop a mixture of sufficient toughness through the introduc- 

 tion of a bituminous binder. In addition to cases heretofore cited 

 where this has been done successfully the utili.ration of the coralline rocks 

 or soft limestone of Florida offers an excellent example of the possibili- 

 ties of such a practice. The majority of these rocks are so soft that a 

 toughness cylinder can not be cut from them, but they have been used 

 in bituminous macadam to form a tough and durable wearing surface when 

 combined with a properly selected bituminous binder. 



As a matter of interest the results obtained with each bituminous 

 material have also been averaged in Table VIII, showing that the fluxed 

 native asphalt gives the highest average result, with the heavy refined 

 tar, Y', a close second. A reference to these averages also shows that 

 for the proportion of bitumen used in these tests the strength of the oil- 

 asphalt specimens decreases with increase in hardness of the binder. It 

 will be noted, however, that, as in the first series of tests, there is no 

 apparent direct relation between any one physical characteristic of the 

 rock and its behavior with a bituminous binder. 



With the object of learning whether any single component of the rocks 

 was responsible for this difference or behavior, the mineral composition 

 of each was tabulated in the same order used in the other tables. The 

 analyses, made in accordance with the method used in the laboratory of 

 the Office of Public Roads and Rural Engineering,^ are given in Table IX. 

 From these results there does not appear to be any definite relation 

 between the mineral composition and the toughness. 



1 Hubbard, Provost, and Jackson, F. H., jr. the results of physicai, tests ok ROAD-BurtDmo 

 ROCK. U. S. Dept. Agr. Bui. 37°. P- "• i9i6. 



2 Lord, E. C. E. relation of mineral composition and rock structure to the physical propBR- 

 tiES op road materials. U.S. Dept. Agr. Bui. 348, 26 p., 3 fig., 8 pi. 1916. 



