On natural and artificial Puzzolanas . 
115 
L—On natural and artificial Puzzolanas. By M. Girard, Engineer 
of Bridges and Highways. 
[From Annales de Cliimie et de Physique, xxxv. 140.] 
The chemical researches undertaken by many distinguished philosophers, to 
discover the cause of the properties possessed by volcanic and artificial puzzolanas, 
have not hitherto ended in giving a tolerable theory of a phenomenon so common 
in the arts : perhaps this proceeds from the circumstances of this phenomenon not 
having been appreciated at their just value. What in fact distinguishes puzzola- 
nas from other earthy substances, is only the property of acquiring a certain de- 
gree of hardness when they are intimately mixed with hydrate of lime, and the 
compound is kept under water for a longer or shorter time. Those are called bad 
puzzolanas which, in the^e circumstances, give a product which never acquires 
more than a very moderate hardness, or rather those which require a month or 
more to become solid. The whole phenomenon requiring explanation consists, 
therefore, as is manifest, in the degree of hardness obtained at the end of a given 
time. Now, it is known that hardness is not in the number of properties w hich, 
in nature, Can generically distinguish a substance. The same quantities of the 
same elements give rise to a multitude of bodies, the hardness of which varies in- 
dehnitely. Thus, from chalk to marble we may mark more than twenty grada- 
tions in the resistance of the almost pure calcareous carbonate. It should, there- 
fore, be expected that analyses, by only proving the quantities of silica, alumina, 
and oxyde of iron contained in clays, would teach nothing, or at least very little, 
respecting their puzzolanie properties. This is confirmed by experience, and it is 
now necessary to seek, in less essential circumstances, ihe causes of these pro- 
perties. 
Chemical facts not having cleared up the question, Messrs. John and Berthier 
appear to agree in attributing the puzzolanie properties to the cohesion and the ab- 
sorbing power alone which the matter acquires by the action of fire. 
But the properties of argillaceous fossil sand or gravel, which I was the first to 
point out, and those which Messrs. Meril and Paven discovered at the same period 
although in a degree weaker, in the grauw r acks and decomposed granites of Bre- 
tagne, do not allow' the opinion of Messrs.' John and Berthier to l>e adopted, at 
least without restriction And I think that I am prepared to establish now, that 
vn fact the cohesion and absorbent power are not, in any degree, the causes of puz- 
zolanic properties. 
M. Vicathas examined ( Ann. de Chim. for June, 1826) what was the influence 
of calcination on each of the elements of a clay which, calcined itself, gave a good 
puzzolana. This inquiry should seem to throw a great deal of light on the theory; 
yet its only result was to prove that silica, separated by acids from unbaked clay, 
is an excellent puzzolana, and loses part of this property by calcination, w r hilst alu- 
mina, which is only a had puzzolana, gains a little by calcination, but too little to 
compensate for the loss of the 'silica ; so that M. Vieat ia led to conclude from this 
experiment, that it is not correct to assimilate what passes in an intimate mixture 
°f silica, alumina, and oxyde of iron, submitted to a weak calcination, with what 
takes place when the same oxydes are calcined separately. The question, then, 
remains entire ; and the following is succinctly the result of the experiments which 
1 have tried with a view to resolve it, and which every body can easily repeat. 
If the clays to which sands or gravels owe their puzzolanie properties be se- 
parated by washing, and the most energetic and the least so are chosen ; if the 
same operation be performed on certain argillaceous sands of the colour of dark 
wine-grounds , which have not, as puzzolanas, any but negative properties ; if to 
these samples are joined pure clays, namely, free from sand, and more or less 
ochreous, such as are found abundantly every w'liere ; if each of these clays, dried in 
the air and powdered, be mixed with half its volume of fat hydrate of lime, and the 
different mortars thus obtained be immersed, in the consistence of a firm paste, tve 
tnall be naturally induced to divide the clays made use of into three classes, calling 
them : 
Good puzzolana clays , those which shall have given mortars which, at the expi- 
ration of ten or fifteen days at most, shall resist the strongest pressure of the finger 
