A very needful branch of recording is the taking impressions of inscriptions and flat reliefs. The usual method is by wet squeezing of paper, which may be made up of any thickness, from a true mould to a slight surface impression. If a mould is wanted for future casting, a tough rag paper without much size should be used; but good newspaper will do. The tougher the paper is when wet, the better. The stone must be thoroughly cleaned and soaked. The paper is cut to the size, and, if less than the stone, in two or more sheets. A sheet is then put in a basin of water, rolled about to soak, and then gathered into a ball and rolled between the hands to break the grain, just short of pulping the surface; next shaken out like a wet handkerchief, and then laid on the stone with enough slack to go into all the hollows. It is then gently beaten with a spoke-brush until it is pushed into the hollows (Fig.33). If they are deep it is needful to use strips of paper soaked and pulped, and laid by finger in the hollows, so as to nearly fill them. Finally, a severe beating is given to the whole, as violent as can be done without tearing the paper. The paper should be pulped on the stone, and driven into every crack and porosity; using a second, and even a third, sheet to bind it together. The pulp in the hollows should be kneaded in with the sharp edge of the brush-back, using the whole weight of the body to force it home. About 50 square feet of such work is as much as can be done in a day. The precautions are: avoid bubbles of water or air below the paper, beat quite straight without dragging, and see that there is no creeping of the paper or shifting on the stone. When quite dry and hard the cast may be carefully peeled off. After heating and waxing, plaster casts may be taken from it, with a slight oiling between each using.

A slighter working is enough on shallow inscriptions; but such squeezes generally need to be taken off while wet, and allowed to dry alone, or else the paper drags flat out of the hollows when contracting in drying. This is specially the case on polished granite, where there is no grip on the surface.

Surface impressions of incised carving may be taken with a single sheet of paper beaten just enough to catch the edges of the cutting; and such make excellent bases for inking over to produce a facsimile drawing (Fig.31). The impression is so much better on the inner side, that the inking is done on that, and the figures are thus reversed in the plate.

But on all coloured work, and many kinds of tender stones, wet squeezing is a crime, as it destroys the original. Fatuous tourists and brazen students have wrecked innumerable monuments by wet squeezing, and it is now necessarily prohibited in Egypt unless special permission is obtained to do some object which cannot be injured by it. Another system, that of dry squeezing, I therefore introduced when doing the Medum tombs. A sheet of thin paper is held over the stone, and it is pressed over each edge of the cutting so as to leave a bend in the surface. Then, laid on a drawing-board, with an oblique lighting, the bends are all drawn on with pencil, checking by comparison with the stone. Sometimes it is best to draw by lamplight, and check with the stone afterwards. The drawing should always begin at the bottom right hand, so as not to press out the impression by the hand; and the sheets must not be rolled before being pencilled. For small lines, a piece of indiarubber should be used to press the paper into the hollows. For the outlines of reliefs the thumb nail must be used. This system is quicker and more accurate than any reduced-scale hand drawing. Over large wall surfaces the sheets should be placed in regular rows, lettered A, B, C (Fig.32), and each sheet numbered in the row, so that A 3, B 3, C 3, come one below the other. The register of positions is kept by marking a minute cross with pencil on the wall, so that the corners of four sheets will fall between the four arms of the cross. Thus each fresh sheet is placed exactly to fit the sheets which have preceded it, in the row and in the column. Any large blanks or injuries should have their corresponding sheets duly lettered (even if nothing is on them), and put with the drawings, so that there shall be no hitch in placing them all in one great sheet afterwards. It may be convenient to join up the sheets, and then redivide the drawings at suitable spaces between the subjects for convenience of packing. To join the sheets they must be laid together in position, a slight cut then made with a knife to mark two sheets across the joint; then turned back-up, adjusted by the cut, and a strip of adhesive paper put on the joint, dabbed down and not rubbed along. Thus large sculptured walls can be copied sheet by sheet, joined up, inked in, and then photolithographed for plates. It is needful to remember that the Postal Union will take rolls up to 60 centimetres length and 21 cm. diameter, as ordinary parcels up to 5 kilograms; or 75 cm. length if not over 10 cm. diameter and 2 kilograms of weight, by book post, open at ends.

Beside the direct material for publishing in plates, it is often desirable to take casts and impressions, both for future reference and also as a step toward a photograph (Fig.34). The making of paper impressions or squeezes has already been noted. Casting with plaster of Paris is the principal mode of reproduction, and is such a detailed business in itself that only a few notes can be given here, such as might possibly be wanted in field work. The fine work for museum purposes is outside of our aim here. The main point in handling wet plaster is rapidity; and for that everything must be ready, and the exact plan of work and amount of plaster settled beforehand. A basin should be used with water equal to about two-thirds of the volume of plaster required. Into this shake or sift dry plaster rapidly, until the water is just filled up with it, and no free water left on the top; it is then well proportioned, and should be violently stirred with a large flat spoon or slip of wood and poured out in an even stream, beginning with the middle if a flat mould, and flattening it out to the edges. It is best to have rather too little than too much; as a fresh lot can be mixed, with the hardened pieces of the first lot, to serve for a backing; the first lot being, of course, spread over the whole face to begin with. Strings, or strips of butter-muslin, should be put through the mass, if it is large, so as to prevent it falling to pieces if broken later on. Excellent casts are made with a thin skin of plaster on a backing of muslin put on a frame; but this requires more skill than plain work. About 10 minutes after casting the back should be scraped down level, or planed with a wide-mouthed hand plane, which is a very useful tool in finishing casts. No cast of any large size should be left without even support for some hours after casting, as it will settle out of shape if strained. Small quantities of plaster are best mixed with a pocket knife in the palm of the hand.

Moulds for casting are usually of clay for a large scale, but that is not likely to be used in the field-work. The division of the clay is best done by bedding threads along the face of the object at the lines required, and then pulling them up to cut the clay. The face of the object requires French chalk (steatite powder) on it to prevent cohesion; oiling or greasing spoils the face of the original. For field-work paper moulds are best, and the preparation of these as wet squeezes has been already described. To fit the squeeze for use as a mould, it should be heated and brushed with melted beeswax on the face, without necessarily soaking it through. Any places that are shiny when cold should be warmed and rubbed with cotton wool, so that the face is the true paper cast. Then slightly oil between each plaster casting, or else the warmth of the setting plaster will make the wax stick to it. Several casts can be taken from one paper, if it is carefully handled in peeling it from the plaster each time. Paper impressions of cylinders are best made with blotting-paper, unrolled wet, and left to dry. To remove the cockling of drying, spread a thin coat of stiff paste on card, and press the paper squeeze lightly on it.

Guttapercha moulds are best if many copies are required. To get a sharp impression in this tough material a preliminary mould should be made, of the right shape, but not sharp on the face. This should be thoroughly cooled in water for an hour or more, and then a small quantity of guttapercha from boiling water should be laid in the hard mould and the object pressed in very rapidly and with maximum pressure. Thus the hot material is forced firmly against every part and takes a brilliant impression. Such moulds are used for electrotyping as well as for plaster work. To produce a smooth face to a lump of hot guttapercha, it should be pulled outwards from the middle to all sides by thumbs and fingers, so as to produce a fresh torn face over the whole upper surface.

Sealing-wax is one of the handiest materials, and is used professionally for all the coin reproductions that are published. Only the best wax is of any use for impressions. It should never be allowed to burn or blaze, nor even to boil, but should be gently heated until a large mass will fall quite readily. The object should be wetted moderately just before impressing. So soon as the wax is tough the object should be lifted slightly to make certain that it has not stuck, and then pressed down again till cold. If it has stuck it must be pulled away at once, and the wax picked off while tough. Sealing-wax casts must be oiled before plaster is put to them; and oil does not soften or deteriorate sealing-wax if left on for years. Beeswax, or, better, the mixture called “dentist’s wax,” makes good impressions, and may be used for moulds.

Tin-foil is most useful for rapid impressions, especially from a fragile or delicate object. The thinnest should be used, such as is wrapped round chocolate. To preserve the form of tin-foil it may be squeezed into place with a back of beeswax, and so form a facing to a wax mould for casting a plaster positive. Or it may be pressed alone (forcing it on with soft indiarubber or cotton wool), and then floated, back up, on water, while blazing sealing-wax is dropped into it to form a backing. This mode is very handy for coin impressions, which will travel safely in this form and look well. For round objects, such as cylinders, a tin-foil impression should be made, beating the foil in with a soft tooth-brush; then the foil is to be uncoiled by rolling it upon wax so that the curve is removed without flattening the impression; it is then ready for a plaster casting, giving a flat cast of the round cylinder. In all cases thin gold-foil would be far better than tin-foil; and such an impression might even be preferred to the original object by some Oriental officials.

Drawing is still the main resource for illustration, although photographic processes occupy so important a place. Hand-work is essential for plans, it is the more useful method for inscriptions, and it is the more convenient method for most small objects. There is generally some interpretation needed, to show details which could not possibly all be visible in one uniform lighting, as in a photograph; and this can only be done by drawing all that can be seen in varying lights and aspects. Another superiority of outline drawings is that they are far more easily looked over and referred to than a much less distinct photograph. And lastly, they cost a third or a quarter of the amount for publication. The proper scope of photographs is stated in the next section.

As drawing is almost always to be reproduced by photolithography, or by zinc block, it is essential to have it entirely in full black and white without any grey or half tones. Hence the contrast should be kept as strong as possible; and only China ink of full blackness should be used for fine lines. In wide, coarse work, as full-sized inscriptions from walls, a common writing-ink evaporated to denser quality may be used. Ebony stain, which some use, has the disadvantage of spreading badly if it chances to be wetted. A smooth, glazy-faced paper is good for fine lines, and does not rag up under the pen. Cardboard is pleasant to use, but is awkward to send by post; whereas paper drawings roll up safely in a tube.

A cardinal rule in drawing is that the finest line should come out to 1/300 inch when the subject is reduced to the plate size. Thus a drawing to be reduced to ? by photolithography should have its finest lines 1/100 inch thick. This line of 1/300 inch is the finest which is safe not to break up in reproducing; and of course it spreads a little in the printing. For very slight shade lines rather thinner lines may be used, as it is no disadvantage if they should break.

It is very desirable to have similar objects all reduced to the same scale. For pottery ? is a convenient reduction; for stone vases ?; for metal tools and small objects ½. The drawings of pottery and stone vases are easiest to do on scales ¼ and ½, as the measured diameters have to be laid off as radii from the axis, needing halving throughout. The further reduction is done when photographing for the lithographs; and it is always best to have such a reduction to ?, if not to ½, of the size of the drawing, in order to make it come out more delicate than the hand-work. A very useful system for recording groups of small objects, especially such as are found together, is to lay them out on a sheet (say double the plate size), as arranged for the plate, and then run a pencil round the outlines, and add as much detail as may be needful to explain the objects; thus a pictorial inventory is made quickly, and is far more useful and easier for reference than any written inventory (Fig.35). The pencil should have the wood split off one side of the lead, and be sharpened by cutting to a chisel-edge on the opposite side. Thus the point is vertically under the guiding side; and when held carefully upright, outlines can well be run from surfaces half an inch or even an inch above the paper. The size of the sheet will, of course, depend on the amount of reduction intended. For numbering the figures printed numbers can be gummed on to the drawing.

For vases, block tints are more satisfactory than outlines. So the drawing can be filled up with a wash of ink. Or if section lines are wanted it is best to draw the section line, and block out the ground outside of the vase, leaving the vase white on a black ground; then have this reversed, black for white, in the photolithographing. The vases may be printed in any colour which is suitable.

The method for drawing a completed form of a vase from fragments is to place the brim and the base (the curves of which can be accurately measured against a series of concentric circles) into their true positions, to a vertical axis; and then, if there is no complete connection, to adjust their height on their axis so that their curvatures (including other pieces which join them) fall into one line. It is easier to do this with the mouth downwards. A frame is made (Fig.36), with a vertical rod sliding up and down over the middle; a card with concentric circles on it is placed on the floor of the frame, and centred under the rod. Taking a piece of a brim, it is rocked to and fro until it touches the card all along the edge, and a leg of wax is stuck on so as to keep it at that angle. It is then slid about till the curve fits between the concentric circles. A piece of base has its curvature measured, by fitting a sheet of celluloid ruled with concentric circles to the curve of it. It is then fixed on the lower end of the vertical rod with some wax, so that the rod is in its axis. Then the rod is slid down in its grooves until the curves of the piece of base and of the piece of brim fall into one line. For drawing the form the radius of the brim and of the base are already measured; the height is taken as it stands in the frame, also the greatest radius at the shoulder, the angle of the side with the base, and sometimes the height from the brim to the curve at several different radii, read off by sliding a graduated square on the concentric circles to touch the curve. After plotting all these dimensions the curve is drawn in by freehand, looking carefully at the fragments in position.

For hand copying inscriptions of a small size, a good method is to fold over the paper at each line that is done, and draw the signs one by one on to the fresh edge of paper held side by side on the stone; thus there is no strain or loss of time by looking to and fro and finding the place, no chance of omissions, and the facsimile is as accurate as possible. This is especially for copying ink writing and graffiti. When making a reduced copy by hand it is best to have a sheet of card under the paper ruled in squares (of ¼ or up to 2 inches), with thick lines. These show through the paper, and a frame of strings or threads is put over the stone, of a larger size, agreeing to the scale of reduction intended; e.g. for reducing a wall to ?, have card ruled in 1 inch squares, and a frame of strings 5 inches apart over the wall. For lines or columns of inscription it does to rule the column lines and only have a long scale on a strip of wood put alongside of the column which is being copied, so as to tally with the lines seen through the paper.