Ochre from—
| Comoal (Savoy). | Vierzen. | St. Georges. | |
| Ferric oxide | 19 | 23·5 | 25 |
| Lime | 2 | — | — |
| Alumina | 20 | 69·5 | 70 |
| Magnesia | 1 | 69·5 | 70 |
| Silica | 44 | 69·5 | 70 |
| Water | 7 | 7 | 5 |
In the majority of cases the mineralogical characteristics of an ochre enable conclusions to be formed as to its suitability as pigment. Good ochre is more or less yellow to dark brown in colour, and can easily be crushed between the fingers to a soft, fine powder which feels like powdered steatite and does not produce a sensation of grittiness, this latter indicating the presence of fine grains of sand in the ferric oxide. The behaviour of the ochre in presence of water is specially important. If it adheres firmly to the tongue, and forms a fairly plastic paste when mixed with a little water, the mineral contains a large percentage of ferric oxide, and as a rule will yield ochre of good colour.
In general it may be said that the value of an ochre varies directly with its content of ferric hydroxide or oxide, because when this is large the ochre will furnish a wide range of colours under suitable treatment.
A simple test for quality consists in weighing out an exact small quantity (10 grms.), and heating it to a temperature not exceeding 110° C., until the weight remains constant. A simple calculation then gives the amount of uncombined water in the sample. Since the proportion of such water varies in different parts of one and the same deposit, the test must be repeated, in order to obtain accurate results, on samples taken from different points, or, preferably, on a properly prepared average sample.
Even drying changes the colour of ochre considerably. To ascertain the behaviour of an ochre on calcination, a large sample is dried at 110° C. until the weight is constant, and divided up into a number of small samples weighing, say, 10 grms. each. The samples are then heated to different temperatures, one to the melting-point of lead, another to that of zinc, and so on.
The higher the temperature employed, the more will the colour of the ochre approximate to red; and specimens very rich in ferric oxide will give bright red colours. Beyond this range, a further increase in temperature will give violet shades, varying with the temperature and the duration of heating. After this preliminary test, it is desirable to make another on a larger scale, with quantities up to about 1 lb. For this test, the different kinds of ochre frequently found in the same deposit should be mixed together, in order to obtain an idea of what the mean product, obtained in working on the large scale, will be like.
On the whole, the results of this second test will be the same as in the first series, the only object of the second test being to gain information which may be particularly valuable in practical work. The bottles in which the calcined samples are stored should be marked with the temperature and length of heating, so that, when it is subsequently desired to obtain an ochre corresponding to a particular sample, all that is necessary will be to heat it to the same degree from the same length of time. The performance of this simple test will be of great assistance in standardising the work with a minimum loss of time.
When it is desired to ascertain the composition of an ochre superficially its behaviour towards hydrochloric acid may be noted. A weighed quantity of the freshly dug (undried) ochre is treated with pure acid, free from iron, which will dissolve out the ferric oxide and lime, leaving clay and quartz sand behind. The presence of lime is indicated by effervescence on contact with the acid; and if there is no effervescence, lime is absent. At the end of several hours the acid is carefully decanted from the undissolved residue which is then stirred up with water, left to subside, and weighed when dry. This method will give the amount of substances, other than ferric oxide and lime, in the sample. These substances usually consist of clay or sand.
For a quantitative determination, a small quantity—usually 1 grm.—is weighed out, treated with a corresponding amount of hydrochloric acid, and the solution filtered into a glass. The residue on the filter is washed with distilled water, the washings being united to the acid solution.