STRENGTH OP CONCRETE. XXXIII. 



were not made as carefully as the 6 inches by 6 inches. In 

 Series II. the 1-2-6 concrete is better than 1-2-5 or 1-2-4, whereas 

 in Series III. the 1-2-4 is better than either of the two others. 

 The 1-3-4 is best in both series. 



Fig. 6, Series III., and Nos. 13 to 18, Series II., show the com- 

 pressive strength of sandstone concrete 2 inches gauge ; here also 

 the cements are not the same. The results show in both series 

 that 1-2-6 concrete is the best, also compared age for age, as far 

 as it is possible to make this comparison, since Series II., was not 

 continued beyond 102 days, the same results are seen. In Fig. 6 

 the 1-3-4 concrete appears the best, and the diagram checks fairly 

 well at 100 days with the results recorded in tests Nos. 13 to 18, 

 Series II. 



Fig. 7, Series III. and Nos. 19 to 24, Series II., do not agree 

 at all, and the apparent differences in the conditions of the tests 

 are not sufficient to account for the difference in the results. 



Fig. 8, Series II., and Nos. 25 to 33 inclusive, do not represent 

 concrete sufficiently similar to compare one with the other as in 

 the foregoing cases. 



In the tests of concrete prisms Series IV., the bluestone was 

 broken to 1 h inch gauge, and the results Nos, 1 to 6 cannot be 

 compared with Figs. 4 and 8, Series III., they are much more 

 comparable with Nos. 1 to 6, Series II., from which it will be 

 seen that Nos. 1 and 2 agree fairly well in giving the compressive 

 strength from 87 to 100 tons per square foot, but Nos. 4, 5, and 6 

 do not agree ; the remaining results are also irregular. 



Conclusions on the transverse tests. — The results of these tests 

 are recorded in the Table Series V., and any irregularity is due 

 to the preparation of the concrete; the method of finishing the 

 specimens in the moulds with true surfaces could not affect the 

 results in transverse, as in the compressive tests. The greatest 

 value of the modulus of rupture obtained was 312 pounds per 

 square inch. 



Conclusions on tensile tests. — From the nature of these tests and 

 methods adopted, the results can only be regarded as a rough 

 indication of the tensile strength of concrete, which is in every 

 case below the real tensile strength. 



3— Sept. 18, 1901. 



