1918] BITRAL ENGINEERING. 891 



tion in the same sample in 24 hours, 0.03 mg. of nitrites will be the lower limit 

 for bad water." 



The cracking and buckling of cement concrete pavements, H. T. Tuthill 

 {Good Roads, n. ser., 14 (1917), No. 20, pp. 255, 256, figs. -;).— Data on thrfee 

 concrete roads paved with 1 : 1.5 : 3 concrete and provided with transverse ex- 

 pansion joints 80 ft. apart are reported. The joint used consisted of a J-in. 

 strip of creosoted yellow pine and the shoulders in all cases were of a sandy 

 loam. 



It was found that " in the case of the pavement built with the coarse aggre- 

 gate of imported crushed stone, there has been no buckling or heaving of the 

 slabs. The pavement developed cracks, but the number of them as compared 

 with the number in the gravel concrete was less, as was also the width of the 

 cracks. No buckling occurred in the pavement built with the deep ditch section, 

 in which the coarse aggi-egate was gravel. There were numerous cracks, how- 

 ever, running in every direction, those occurring longitudinally of the pavement 

 opening very wide. On the shallow ditch section of the gravel concrete pave- 

 ment, bad ca.ses of buckling, considerable distances apart, occurred within a 

 week during an extremely hot spell. Longitudinal cracks were also more or 

 less frequent on this section, but tliere were few diagonal cracks." 



Tests show advantages of laying brick directly on concrete base, C. C. 

 Wiley (Engin. Netcs-Rec, 79 (1017), No. 18, pp. 820-822, figs. 5).— Tests at the 

 University of Illinois are reported on 66 slabs, including plain concrete slabs in- 

 dented to represent ordinary concrete roads and monolithic brick slabs (1) con- 

 sisting of grouted brick only, (2) with a thin layer of mortar used to smooth 

 the surface of the concrete, and (3) with the bricks laid directly on the green 

 concrete. Wire-cut-lug bricks were used in virtually all slabs. 



" Comparison of the results given in the accompanying tabulation will show 

 that when the concrete side of the slab is in tension the direction of the brick 

 courses has little effect on the strength, and that the modulus of rupture is 

 approximately that of plain concrete of the same quality. This is true when the 

 concrete forms roughly one-half of the thickness of the slab. With the very 

 thin bases, the brick makes up a sufficient part of the total slab to govern its 

 behavior. 



" When the bricks are in tension there is a marked difference in strength 

 with the direction of the courses of brick, as would be expected. Here the 

 slabs with longitudinal courses show high strength, in most instances greater 

 than that of 1 : 2 : 3 concrete. This is doubtless due to the fact that the bricks 

 lap in such a way that a shearing stress is set up between the bricks and the 

 grout, and the bond in this direction is sufficient to develop high beam strength. 

 The slabs with transverse joints are very much weaker since in this case the 

 stress is direct tension on the bond between the brick and the grout. The 

 beams with diagonal courses show a strength intermediate between these 

 two. . . . 



" Little difference in strength is noted between the ' dry-layer ' type and the 

 ' direct-contact ' type of construction, and what difference exists is probably 

 due to a difference in the gi-out. The * dry-layer ' slabs used a grout of J-in. 

 sand, which a few tensile tests showed to yield a stronger groitt than the i^-in. 

 sand used in the ' direct-contact 'specimens. The grout of i-in, sand was 

 decidedly more difficult to mix properly and showed more tendency to segregate 

 than did the other — a point worth considering in actual construction. In the 

 ' dry-layer ' type of construction the bricks show the greater tendency to sepa- 

 rate from the concrete. The materials in a dry mix are never associated as 

 intimately as those in a wet mix, and consequently the cement does not come 



