XIII, A. 4 Witt and Reyes: Solubility of Portland Cement 159 



that the liquid passing the membrane contained aluminium and 

 calcium in about the original proportions. They conclude from 

 this that no colloid is formed and that the substance is not broken 

 up by hydration. 



The work in this laboratory favors Stern's results, though it 

 must be remembered that commercial cement was used in every 

 case and not an aluminate alone. It may be also said that 

 if a colloid forms according to Michaelis's theory it is broken 

 up by a large excess of water, as the presence of such a large 

 amount of dissolved calcium with only a trace of silicon (in any 

 form) shows. Or the explanation may be that the colloid does 

 not form because the concentration of the calcium hydroxide 

 solution is not sufficiently high.-^ 



It is generally conceded by cement investigators that the 

 strength of a test specimen depends to some extent on the 

 fineness of grinding; in fact there is no doubt that, other 

 factors being equal, the finer a cement is ground the greater 

 strength it will give mortar briquettes. A proof of this is 

 that if specimens of hardened mortar or paste are reground 

 the powder may be again mixed with water, and a fair 

 degree of strength obtained.-- The mass may be again ground, 

 and water added, with a like result. The usual explana- 

 tion offered for this is that during the first gauging the water 

 cannot penetrate the larger particles of cement and that the 

 cores of these remain unchanged. When reground and regauged, 

 these parts become active, and there is sufficient new paste 

 to cement the whole together and so on. The present work 

 supports this explanation, but indicates that there are other 

 factors to be considered. By referring to Tables IV and VII 

 it may be seen that, although only about 40 per cent of the 

 total calcium in the cements was hydrated and dissolved when 

 agitated for fifteen days in the original experiment with fine 

 cement, approximately 90 per cent of the calcium went into 

 solution in only twenty-four hours, when the relative volume 

 of water was increased twentyfold. Since the cement was of 

 the same fineness in both cases, it may be seen that the volume 

 of water is of importance as well as the size of the particles. 

 Further it is probable that if the finest cement flour obtainable 

 were gauged with water it would not be completely hydrated — 

 not because of size of grain, but because of reasons already ex- 



" Michaelis, loc. cit. 

 " Michaelis, loc. cit. 



