IOWA ACADEMY OP SCIENCE 
113 
land Cement, this would show that an average of over eleven per cent of the 
Portland Cement had been dissolved and removed by the extractor. The piece 
-of tile No. 2 was again subjected to the Soxhlet extractor for 72 hours and 
the amount of solids secured was 0.4498. This time the percentage amounting 
to 2.55 per cent. 
Mr. Layton Gouldin, assistant to Dr. Knight, has made analysis of these solids 
and finds the percentages to be; Ferric iron, 0.97 per cent; Alumina, 2.08 per 
cent; Magnesia, 0.21 per cent. The remainder of 96.74 per cent being calcium 
carbonate, silica and calcium sulphate. The work is not yet complete on the 
analysis, but the qualitative tests show the presence of the elements not de- 
termined. 
Mr. Layton Gouldin has questioned the reliability of work done on a sample 
of the concrete tile as being open to the suspicion that contamination might 
result from the sand and gravel which forms four-fifths of the body of the tile. 
He has, therefore, made a number of preliminary tests upon various samples of 
neat cement, containing no grog whatever. The results of these tests, which I 
cannot give in detail, show clearly that the Portland Cement is by far the most 
soluble part of the concrete and that we are justified in believing that the ele- 
ments of the gravel are practically insoluble, as compared with the Portland 
Cement. 
Another line of approach to a solution of this question has been to take 
cement that has not been set and subject it to the action of distilled water. 
These two fiasks contain 5 grams each of standard high grade Portland Cement 
in 200 cc of distilled water. You will notice that the cement has passed en- 
tirely into solution or remains in suspension. Work at the University of Illinois 
by Mr. L. R. Ernest warrants us in believing that the water has the power to 
dissociate the elements of the cement into calcium oxide, aluminum hydroxide 
and silicic acid. You will see that the Portland Cement five grams in 200 cc of 
water has swelled to more than thirty times its original bulk. We further 
treated pulverized samples of hydrated Portland Cement with distilled water 
in like manner. While the results are not nearly so marked in the case of the 
hydrated cement as they are in the former, the results parallel the first results, 
but are much slower. 
At the end of thirty days the hydrated cement has swelled to more than eight 
times its original bulk. This swelling of bulk and practical dissociation of the 
elements is slowly continuing, showing that the process of hydration or set- 
ting when completed has not removed cement beyond the power of water to 
dissociate its elements. In these tests, while not a complete solution can be 
accomplished, the very marked dissociation of the elements, amounts to the same 
results practically in the destruction of concrete work as though the silica, 
aluminum and calcium were entirely soluble and would pass through the filter 
paper. In connection with these experiments and our other work, all of which 
show the power of water, either to dissolve or dissociate the elements of con- 
crete, we wish to spend a brief moment in theorizing. If water will dissolve or 
dissociate any considerable part of the Portland Cement, then it will necessarily 
follow that perfectly hydrated and completely set Portland Cement can be in- 
duced to reset and develop a very appreciable degree of tensile strength by 
pulverizing it finely and subjecting it to conditions similar to those under which 
it was first set. We state, therefore, today, that it will be found to be true that 
Portland Cement, thoroughly hydrated and completely set, can be pulverized 
