324 



Journal of Agricultural Research 



Vol. X. No. 7 



Table VI. — Effect on toughness of variation of rational proportion of bitumen and sand 



Test No. 



Toughness. 



I. 



2. 



3- 

 4- 

 5- 

 6. 



7- 

 8. 



9- 

 10 

 II 

 12 

 13 



3-3-3 

 4-4-4 



6-6-6 



7-7-7 

 8-8-8 



lO-IO-IO 



II-II-II 



II-IO-II 

 lO-IO-IO 



9-9-9 



8-8-8 

 6-6-6 



S-5-5 



For the more complete information of the reader Table VII gives the 

 percentage by weight of each bituminous material used with each rock. 

 In this and all other tables the rocks are arranged in the order of increas- 

 ing toughness with the light refined water-gas tar at the end of 24 hours. 

 The results obtained in this series are given in Table VIII. While, as above 

 noted, they are uniformly lower than those obtained in the first series, it 

 will be seen that the various rocks bear the same general relation to each 

 other, as shown in Table IV, for the first series of tests. With the light 

 water-gas tar X', for instance, the diabase and quartzite continue to 

 exhibit no strength whatever, while the chlorite gneiss shows up well 

 from a strength standpoint. This bears out fully the observations made 

 in the actual construction work which led to this investigation, where the 

 bulk of the work referred to was constructed with these three types of 

 rock, and the chlorite gneiss was the only one which showed any pro- 

 nounced success in combination with the relatively fluid tar binders. It 

 also may be noted with interest that the feldspathic sandstone 8136 

 shows a marked weakness in combination with all the binders, which is 

 clearly shown in the right-hand column of averages of all results obtained 

 on each type of rock. This accords with some results obtained recently 

 in actual practice on experimental sections constructed by this office. 

 The same rock from which a sample was taken for these tests was used in 

 a one-size stone bituminous concrete with both oil asphalt and fluxed 

 native asphalt of 102 and 117 penetration, respectively. The section 

 began to check and crack almost immediately, and in a few months began 

 to fail so generally that a surface treatment was necessary in order to 

 save it. 



