SOME NOTES ON ANCIENT MOETAES. 37 



surfaces of the stones or bricks, and prevent the lime from 

 unduly shrinking as it hardens. 



The best material is undoubtedly quartz sand, but not too 

 fine. If the sand is very fine, the mortar becomes too dense 

 to admit the air which is needful for its solidification ; and 

 if it be too coarse, the interstices become unduly large to be 

 filled with the lime. A mixture of coarse and fine sand has 

 been often employed, and angular sand has been much pre- 

 ferred to the smooth round sand. 



Chemists have differed considerably on the cause of the 

 setting of mortars, and from time to time various theories 

 have been raised. At one time the hardening w^as con- 

 sidered due to mechanical agency, and now it is thought 

 that chemical combination contributes to render it durable. 

 This chemical action, says an author in a well-known book,^ 

 is not wholly confined to the formation of carbonate of lime. 

 The action which takes place during the hardening of mor- 

 tar appears to be somewhat as follows :— The carbonic acid 

 of the atmosphere, acting upon the exposed parts of the 

 mortar, forms a coating of calcium carbonate ; the amount of 

 this substance formed increases with the age of the mortar; 

 but at no time, so far as we can learn, is the whole of the 

 lime of the mortar converted into carbonate. Besides cal- 

 cium carbonate there is also formed a small quantity of cal- 

 cium silicate, by the action of the lime upon the silica of the 

 sand and of the stone or brick ; this silicate spreads over 

 the surface of each little grain of sand, and binds the whole 

 mass compactly together. The formation of the carbonate 

 continues so long as the mortar retains moisture to dissolve 

 the lime, and the air has access. Whatever may be the ad- 

 vantages gained from these combinations, they are not sufii- 

 cient to impart firmness to the material ; it is only from the 

 1 " Chemistry as Applied to the Arts and Manufactures," vol. ii. p. 460. 



