142 W. H. WARREN. 



The adhesive resistance was always much below the 

 elastic limit of the material, as it was impossible to provide 

 the necessary adhesion area between the fractures and 

 the extremity of the head of the specimen. 



If I = the length of the rod required to provide an area 

 sufficient to develop the elastic limit of the material, 

 c = circumference of the rod. 

 p g — the adhesive strength per square inch. 

 f — the elastic limit of the material per square inch. 



Then: } _ Ja^ 



If the diameter of the steel rod is \ inch and circumference 



= 1*57 inch 



j _ 42000 x Q'1963 = 5251 



l"57p a p s 



The value of /> s varies with the age of the mortar or con- 

 crete, and with the quantity of water used in mixing. 



It will be seen that it varies considerably in the various 

 tests recorded in Tables I. and II. 



If p B = 100, then I = 52*251 inch. 



With a length of 7 inches which represents the distance 

 from the fracture to the extremity of the head in a large 

 number of tests, the adhesive resistance would have to be 

 750 lbs. per square inch to develop the strength of the 

 metal at the elastic limit. 



In I. b. 1 to I. f. 6 the same shackles were used to hold 

 the heads, but the specimen was made longer and reduced 

 to 4 x 3 and 3x3 for the middle 4 inches, as shown in 

 the sketch on Table I. b. 1, with a view to provide a larger 

 adhesive resistance and develop a greater tensile strength. 

 The largest stress sustained was in I. b. 5, which gave 1611 

 pounds per square inch, although the mortar w r as only 42 

 days old. The strength of this mortar w r as certainly not 

 greater than 250 lbs. per square inch, so that a resistance 



