manganese oxide. The character of the base varies according to the locality where the pigment is found. The analyses given below illustrate this point very well.

As stated above, these pigments owe their colour to hydrated peroxide of iron; so far as can be judged from the composition and manner in which they are found, this oxide of iron is, in all cases, an alteration-product, formed by the oxidation of ferruginous minerals, especially pyrites, and, in the case of siennas, of manganiferous ores. The manner of occurrence of ochres and siennas, so far as information at hand is available, appears to be different; in the case of ochres, these seem to have been formed by the decomposition and oxidation of iron minerals at the spot where the pigment is found, for, in the majority of cases, the ochre occurs more or less mixed with the gangue or matrix, from which mineral it originated. The analyses of the Welsh and Derbyshire crude ochres given below are examples of ochres formed in this manner. This occurrence of much matrix or gangue in crude ochres necessitates the adoption of a method of levigation to prepare them for use as pigments.

Siennas occur under various conditions, generally as aqueous deposits, in hollows and basins; some of these are of recent formation, since bronze idols and other articles of human manufacture have been found in them. Siennas seem to have been formed by water draining over beds of iron and manganese ores, so as to become charged with oxide of iron and organic matter, and then flowing into a hollow in which the iron, &c., it contained was deposited; in process of time the hollow was filled up and the bed of sienna formed. Siennas differ from ochres in containing very little extraneous matter, and, therefore, they require little beyond grinding to fit them for painting, when they are known as "raw sienna" (see p. 149). Ochres are prepared for use as pigments by grinding and levigating. The plant used for this purpose varies at different works, its construction being largely dependent upon the nature of the ochre which is being treated. Some ochres are soft and powdery, these only require levigation; while others are harder, and need to be ground before they can be levigated. Fig. 28 shows one form of this plant, and in Fig. 12 is shown another form used in the china-clay industry, which may also be used for ochres. A, Fig. 28, is a large tank which, in Derbyshire and Cornwall, is known as the "buddle"; in this is placed the rough ochre, just as it comes from the ground, if the nature of the material permits; or it may be subjected to a preliminary grinding in an edge runner mill. In the buddle it is

thoroughly mixed with a constantly flowing gentle current of water, which carries off the finer particles of the ochre, and leaves the very coarse material behind; as the latter accumulates in the buddle it is removed from time to time and thrown away. From the buddle the water carries the ochre into a settling-tank, placed at a rather lower level, where the coarser particles of ochre settle out; when they have accumulated sufficiently the water is run out of the tank, and the ochre removed. From the first settling-tank the water, which still contains some ochre, is run into a second tank rather larger than the first, so that the current of water becomes more sluggish, and the finer particles of ochre subside more readily; in this second tank the ochre is allowed to settle until enough has accumulated, when it is removed as before. From the second tank the water runs off into a third tank, in which a still finer quality of ochre settles out. All these tanks are shown in Fig. 28; sometimes a fourth tank is arranged. The number of settling-tanks much depends upon the quality of crude material and the variety of grades of ochre it is desired to make; the grade of ochre which settles out in each tank is different from the others; that in the first is darker in tint and coarser in quality than that which settles out in the third; while, if a fourth tank is used, this will give a pale and fine quality of ochre. It will be found best in large works to arrange for three sets of settling-tanks, arranged side by side; the material from the buddle is run into the first of these sets until the tanks are full; then the stream is turned into the second set, which are also filled; then the stream is turned into the third set; by this time the ochre in the first set will have settled out, and the tanks can be emptied and got ready for the water to be turned into them. When the third set has become full the second set will be ready for emptying, and this is done; then the third set is emptied; thus there is always one set being filled, another settling out, and the third being emptied. There is scarcely two ochre works where the arrangement of the levigating plant is alike, as much depends upon the locality, the amount of water at disposal, and the quantity of material being treated.

After being taken out of the settling tanks the ochre is in the form of a paste which, as it contains probably 50 per cent. of water, must be dried. The arrangements for this vary very much at different works; at some, long horizontal flues are built with a fireplace at one end, and the flues covered with flagstones on which the wet ochre is placed; at others, the ochre is placed

in earthenware pans in an ordinary drying stove; but in every case, the drying of the ochre at as low a temperature as possible is important, as too high a temperature turns the colour of the ochre rather reddish.

PROPERTIES OF OCHRES AND SIENNAS. So far as their properties as pigments are concerned, the ochres and siennas rank among the most permanent pigments at the disposal of the painter. They are unaffected by admixture with any other pigments, do not act injuriously upon other pigments, and are scarcely affected by exposure to the atmosphere and its destructive influences. They work well with all kinds of vehicles, and can, therefore, be used in any kind of painting-oil, water, distemper, fresco, &c.

Ochres and siennas vary very much in tint, brightness of colour, and strength. Oxford ochre is the brightest of the ochres and is of a fairly bright brownish-yellow colour. Siennas are of a brownish-yellow colour varying much in depth of tint or shade. Welsh ochres are rather duller than Oxford ochres; French ochres are moderately bright; Derbyshire ochres are reddish in tone and are darker than other varieties of ochre. They vary very much in texture; Oxford ochre and the siennas are of a soft texture; some are gritty in feel, while others have a clayey feel. In body or opacity these pigments vary much. The Oxford ochre and the siennas are rather transparent, and are commonly used as glazing colours; the other ochres are more opaque and have good body; hence, they are largely used as body colours, especially in house painting.

The colour of ochres is due to the presence of hydrated peroxide of iron, while siennas also contain small quantities of manganese; the shade or tint depends mainly upon the proportion of iron and manganese present, and also, but to a less extent, upon the degree of hydration of the oxide of iron; in proportion as the iron oxide is less and the hydration greater, the yellower and brighter the shade of colour; when the proportion of nonhydrated oxide of iron is large the shade becomes redder. When ochres are treated with hydrochloric acid, the iron they contain is nearly all dissolved out, and yields a yellow solution which will give the characteristic tests for iron, while a more or less insoluble residue is left behind.

Heat turns ochres a red colour, the shade of which depends upon the temperature and length of time the heating is carried on; these red colours are sold as Venetian red, light red, Indian red, &c.; their preparation and properties have already been described (see p. 112). Siennas are converted by heat into a

reddish-orange pigment, known as burnt sienna (see p. 150). This change of colour is due to the passage of the iron oxide from the hydrated to the anhydrous condition, but the reason why ochres should give reds and the siennas orange is not known.

ASSAY AND ANALYSIS OF OCHRES AND SIENNAS. (a) Crude Ochres and Siennas.-These should be assayed for, first, the actual quantity of colour present and, second, for the tint or shade of the colour it gives. This last can be done in the usual way; the first can be ascertained as follows:A tall glass of a conical shape is provided; a glass funnel with a long stem passes down to the bottom of the glass into which is put about 25 to 30 grammes of the crude ochre; into the glass is now passed a gentle current of water sufficiently strong to carry out of the glass all the finer particles of colour while leaving the heavier and more gritty particles behind, which are collected by filtering and, after drying, are weighed in the usual way; from the weight is calculated the proportion of colour and grit. Thus, the sample of crude Irish ochre (an analysis of which is given below) assayed in this way, was found to contain—

(b) Prepared Ochres.-These only need assaying for colour and covering power by the usual methods.

It is rarely that an analysis of ochres and siennas is required; but analyses of several varieties are given below, which show their constituents and what to look for in analysing them. Ochres are rarely, if ever, adulterated. Ochres which are naturally poor in colour sometimes have a little chrome-yellow added to them to bring up the tint; such an addition may be recognised by treating the ochre with hydrochloric acid and alcohol, when a green-coloured solution containing chromium will be obtained, and the chromium in which may be detected by the usual tests.

COMPOSITION OF OCHRES AND SIENNAS.-Most of the following analyses have been made by the author; others are quoted from various published analyses. The notes appended to some of them will be found useful and of interest as showing some indications of the origin of these pigments.

1. Oxford Ochre. The ochres from Oxfordshire have long had a reputation for their quality, exceeding, as they do, all other ochres in the brightness of their colour and depth of covering power. Most of the ochre is found in pits at Shotover, near

Oxford, of which the following section is given in "Ure's
Dictionary" :-

1. Summit of hill, highly ferruginous grit,
2. Grey sand,

3. Ferruginous concretions,

4. Yellow sand,

5. Cream-coloured loam,
6. Ochre,

Oxford ochre contains

3

6 feet.

6 inches.

The layer of ferruginous concretions is, probably, the source of the colouring matter of this ochre, while the clay which underlies the layer of ochre is the source of the base of the pig

ment.

2. Welsh Crude Ochre.—The exact locality from whence this sample was derived is not known; it has a fairly good colour and covering power. It contains

Some of the iron oxide, 0·765 per cent., exists in a soluble form, probably as sulphate, for there is 0.555 per cent. of soluble sulphur trioxide. This, and the fact that there is copper sulphide present, indicate that this ochre has been formed by the decomposition of a cupreous pyrites, which supposition is further strengthened by the fact that small pieces of pyrites may be picked out of the crude ochre. This ochre requires well levigating to get rid of the pyrites as this body would introduce