The strictly chemical view of this subject is beyond the scope of these Notes; but in order to render them more complete, the following directions 3 for testing and analysing a lime or cement are added, in the hope that they may be useful to such readers as may have the chemical knowledge necessary to understand them and put them into practice.

Chemical Test

"A useful chemical test for the amount of hydraulicity is made as follows : - A small portion of the lime or cement (about as much as can be piled upon a sixpenny piece) is placed in a test-tube, and sufficient H20 added to cover it, and finally about l1/2 drachm of HCL. It is then warmed over a gas jet or spirit lamp. If much soluble silicic acid be present it will form a thick jelly, whilst if poor or if it have become inert, it will be only partially gelatinised, and perhaps remain liquid. Taking a cement which is known to be rich in silica, and comparing it with the one which is to be tested, a very fair estimation of its hydraulicity can be made. Limes which contain much magnesia are apt to part with their silica in an amorphous powdery form, yielding only a slightly gelatinous appearance. This should be noted, and a sample not rejected if the 'precipitated silica be sufficiently large in amount. If the cement or lime effervesce very strongly on the appli cation of the HC1, there is a strong suspicion that it has absorbed too much carbonic acid, and has thereby weakened its setting and lasting properties."

1 Burnell On Limes and Cements. 2 Gillmore On Limes and Cements.

3 From Notes on the Chemistry of Building Materials, by Captain Abney, R.E., F.R.S.

Chemical Analysis

"The chemical analysis of a lime or cement is one which is easily performed, and requires but little apparatus.

"About 30 grains of the cement should be powdered in a mortar and placed in a small porcelain dish, and about 2 drachms of H20 and the same quantity of HC1 added to it, and the dish placed on the sand bath; a brisk effervescence being noted, shows the presence of carbonic acid. The powder, after stirring with a glass rod, will begin to dissolve, and after five minutes' boiling a residue will be left which is plainly unattackable by the acid. This must be filtered out, and noted as insoluble residue, consisting most probably of sand. After filtering and washing out the dish the liquid is evaporated to dryness. When the dish is sufficiently cool, a little strong HC1 is poured on the mass, and then hot water, and the whole again boiled. It will now be found that a white residue is left undissolved. This is silicic acid, which must be filtered out.

"A small portion of the liquid is poured into a test-tube, and a little H2S added Should there be no ppt, the absence of Group II. may be inferred.1 Another portion is treated with NH4C1 + NH4H0; this will be found to yield, in almost all cases, a reddish gelatinous ppt. Such being the case, the same re-agents must be added to the whole remaining portion of the liquid which should be transferred to a glass beaker. The ppt must next be filtered out, and well washed with warm H20, and the filtrate (a) left for further examination.

"The ppt being dissolved off the filter by passing a few drops of strong HC1 through it, and the solution being diluted with a little water, KHO is added in excess. A red gelatinous ppt at once becomes apparent. This is filtered out and tested for iron by (NH4)2S. Blackening shows the presence of iron. The filtered solution is now treated with NH4C1, and should a white gelatinous ppt be formed after standing, alumina is present. Solution (a) is next examined, - a small portion of it may be tested with ammonium hydro-sulphate, usually without any result: but should a ppt be produced, it will indicate Mn or Zn, or both; the remaining portion is treated with (NH4)2C03; a dense white ppt will be at once formed, this must be filtered out, and the solution () put away for subsequent examination. The ppt is then treated with a very little dilute HC1, which dissolves it. To a portion of its solution a little CaS04 is added; a ppt on boiling indicates the presence of Ba or Sr, while a ppt caused by (NH4)2C2O4+NH4HO - in a fresh portion of the solution -shows the presence of lime.

1 Group II. consists of metals, in solutions of which HC1 gives no precipitate, but HC1 + H2S give precipitates - the metals are Hg. Cu. Bi. Cd. As. Sb. Sn. The metals of this group are but seldom met with in lines or cements; should any be present, recourse must be had to the tables.

"Solution () is next treated with a few drops of NH4HO and Na2HP04, and left standing an hour; a ppt being then visible gives evidence of magnesia.

"About 100 grains of the cement should be boiled in water to extract any soluble salts, and the filtered solution evaporated down to small bulk; none of the soluble salts of the metals of groups ii., iii., and iv. would be likely to occur in a lime or cement, and consequently the metals of these groups need not be looked for. A small quantity of lime will, however, be taken up by the water. A small quantity of ammonia, and ammonium oxalate, and an excess of ammonia, must be added to the solution to precipitate it, after which the filtered solution should be evaporated to dryness. The residue should then be strongly ignited to drive off the excess of ammonium oxalate, and should there still be a residue it will probably consist of alkaline chlorides present in the cement. To confirm this result, redissolve the residue in water and test a portion for chlorine with silver nitrate, which should produce a white curdy precipitate, insoluble in boiling nitric acid.

"Finally, if any soluble sulphate be present it will have been taken up by the H20; therefore, a fresh portion of the solution should be tested with barium chloride, which will produce a white precipitate if any sulphuric acid be present.

"A further search for acids is unnecessary, no one of any importance or of any value could be found, as any such would have been traced in the examination for metals. The qualitative examination of a lime or cement usually gives the following as present: -

"Carbonic Acid. Insoluble Residue. Silicic Acid. Ferric Oxide.

Alumina.

Lime

Magnesia.

Alkalies. Chlorine. Sulphuric Acid (trace)."

The following Table gives analyses of a few cements (made by a friend of the writer's) taken from actual specimens of fair quality met with in practice :-

Portland Cement.

Roman Cement.

Medina Cement.

Heavy slow-setting.

Light quick-setting.1

Average good cement.

Specimen 1.

Specimen 2.

Clay unacted upon ..

•3

4.4

Traces.

97

7.9

53

Soluble Silica ...

22.3

20.6

22.0

16.0

17.2

19.0

Oxide of Iron2 ...

3.2

3.5

3.5

22.2

21.5

16.6

Soluble Alumina..

7.2

10.9

8.0

Sulphuric Acid..

1.4

2.4

1.0

Lime ...

63.0

50.0

62..0

41.2

46.1

49.8

Magnesia ....

•6

•2

1.0

1.7

1.6

Alkalies ....

1.3

1.5

1..5

9.2

5.7

9.3

Carbonic Acid ...

5.0

Traces.

Moisture and loss..

7

1.5

1.0

1 This cement evidently contains too little lime. 2 Ferric oxide.