244 REIBLING AND REYES. 



According to this diagram, the slaking of the free lime in a 3.5 per 

 cent mixture would only be capable of retarding the time of initial set 

 from one of 32 to one of 45 minutes, whereas the same action in a 2.0 

 per cent mixture would be capable of changing it from 25 to 90 minutes. 

 Therefore, it is evident, as actual experience proved, that a slower setting 

 cement could be obtained by seasoning a 2.0 per cent mixture. 



This material also could be plastered to insure the requirements 

 specified for abnormally quick-setting cements, as it is now evident that 

 the gi-ound, weathered clinker if mixed with 3, or better 5, per cent of 

 plaster would remain abnormally quick setting unless subjected to such 

 drastic seasoning as to render the cement practically inert. 



The results obtained with this sample again indicate that it is char- 

 acteristic of the general nature of the commercial product which it 

 represents. Take for instance the representative commercial product 

 recorded in Table A, Appendix I.^** It was plastered with an equivalent 

 of 2.7 per cent (1.44 per cent of SO3, ignited) of our plaster of Paris, 

 set in 10 minutes and remained quick setting during three months of 

 thorough aeration. 



The reason why these three cements behave so differently is because 

 the rate of the reaction of the compounds which cause them to set are 

 not the same. It can readily be realized that the present and future 

 setting properties of a cement depend not only on the quantity and the 

 condition of the free lime and the amount of retarder capable of entering 

 into the reactions involved, but also upon the quantity, character and 

 activity of the setting compounds themselves. Accordingly, in brand 

 L and similar cements we may assume that extremely quick-setting 

 compounds (presumably the dicalcium aluminates and ferrates) are 

 lacking to such an extent that in spite of changes in the condition of the 

 free lime and variations in conditions of heat, a small quantity of 

 retarder is- at all times capable of keeping the set within normal limits. 



We may also assume that the efEect of heat upon the activity of the 

 quick-setting compounds is greater in some cements than in others, and 

 hence the difference between and K. 



We have not encountered cements in which the normal activity of the 

 compounds causing the set was so great that no process of seasoning or 

 plastering was able to produce a normal set, but such cements have 

 frequently been reported from other sources. 



However, whatever the influence of chemical composition, the real 

 phenomena entering into the reactions, the natural activity of the setting 

 compounds, and the quantity and condition of the free lime, it is still 

 evident that the method of investigation described in Part III of this 

 paper is an accurate and efficient means of ascertaining the possible 



^^TMs Journal, Sec. A (1910), 5, 415. 



