240 REIBLING AND REYES. 



must give special consideration; and therefore, if he packs his cement 

 in sound, paper-lined barrels or in air-tight, water-proof bags, he need 

 only concern himself with the changes in set which are brought about 

 by the conversion of free lime into its hydrate. The free lime in Port- 

 land cement packed in this manner and properly stored can not possibly 

 become converted into calcium carbonate except to a slight extent.^" 



For example, these same mixtures of Oi cement were also stored, packed to a 

 depth of 12 centimeters in open beakers. The only changes in set which took 

 place were those caused by the hydration of the free lime: i. e., the mixture 

 containing 0.5 per cent of plaster, after first becoming quick setting, became and 

 remained slow setting; while the other mixture remained slow setting at all times. 

 The microscopic test showed that only the cement on and near the upper surfaces 

 became thoroughly aerated and that the free lime in at least four-fifths of the 

 cement did not become converted into calcium carbonate. Consequently, only 

 the effects caused by the hydration of the free lime manifested themselves in the 

 combined results obtained. (See Table XX.) 



The conversion of slaked lime into calcium carbonate is very apt to 

 occur to a considerable extent if the cement is packed in ordinary paper 

 or cloth bags, or if the samples taken from the commercial packages 

 are not properly protected from the atmosphere. However, as this matter 

 has already thorouglily been discussed in one of our pre\'ious publica- 

 tions ^^ it only is necessary further to state that proper packing is es- 

 sential not only for the control of the set but also for the preservation 

 of the strength of good Portland cement. 



It seems that the setting properties of 0^ cement are truly characteris- 

 tic of the general nature of the commercial product which it represents. 

 (See Table XXIII.) 



Experiments with the abnormally quick-setting cements received from 

 this manufacturer have resulted in the following observations: 



1. The set of the cements could always be retarded to normal reqmrements by 

 the addition of from 0.5 to 1.0 per cent of plaster. 



2. When aerated, the most lightly plastered cements remained quick setting. 



3. When aerated, the most heavily plastered cements soon became normally 

 slow setting. 



It was claimed (and a later inspection of the records at the factory upheld 

 this) that certain of these quick-setting cements had been carefully tested before 

 they left the mill and that they then were slow setting. At that time we were 

 unable to account for the change in set which had evidently occurred after the 

 cements were packed. However, it now appears that the results obtained at the 

 factory were those produced by partially regaiiged cement pastes. 



The physical testing at cement plants is usually intrusted to em- 

 ployees who although conscientious and faithful, do the work in a purely 



" See Tables IX and X, Part II of this paper. 

 "This Journal, Sec. A (1908), 3, 137. 



