A large site, such as that of a temple or a town, may be attacked in several ways. The most cursory method is by trial pits in various spots; pits which, if they hit anything of importance, are likely to injure it, and certain to destroy its connection with other things. French explorers have a love for faire quelques sondages, a proceeding which often ruins a site for systematic work, and which never shows the meaning of the positions or the nature of the plan. If it is quite uncertain whether there be remains in the ground, the best examination is by parallel trenches, as such give a good view of the soil, while the stuff can be turned back and the trench filled behind if not wanted. In case of tracing a building, trenches cut along the lines of the walls are a good beginning; and then if more is wanted, the plan is clear and the rooms can be emptied with foresight.

A favourite method with the older explorers was to clear out a whole area (Fig.18) and throw the stuff all round the site. This may be needful in case of superimposed buildings, which must be studied one by one, as only two or three periods can be planned at once, and the upper have to be removed before the lower can be cleared. But such a method is a clumsy waste in dealing with a simple group of buildings. The great difficulty of it is to know where to place the stuff removed, so as not to block future work. Before beginning any large excavation, the amount to be shifted should be gauged, and the position of the stuff settled beforehand. The great clearance on the side of the Medum pyramid, to expose the temple, was planned out with the position and size of each waste heap in the mind’s eye, and the system of paths by which the stuff could be shifted with least fatigue. It is needful to continually adjust the moving, so as to avoid lifting the stuff more than really needed; and any long run down of material, either towards the digger or away from the thrower, should be prevented, as it all has to be lifted again in some shape. Working at the foot of a long run of stuff is entirely wrong; such ground should be shifted in successive levels, each level being discharged without needing to raise the earth up again. Excavations at the Sphinx were carried on by the Government with two men filling baskets with sand, which ran down 20 feet from the surface to the bottom of a pit; and the baskets were then carried up by a long train of children very slowly climbing up out of the pit on a sand slope at the angle of running sand. Thus nearly the whole labour was wasted by not filling the baskets at the surface and carrying them directly away. Whenever a large pit is needed it should be begun of full size, and lowered equally all over, so that nothing runs down during the work.

For moving earth to a distance there is no way so simple and adaptable as a line of carrier boys (Fig.22). Over flat ground this is the best way up to distances of 50 or 100 yards; for longer discharges it may be better to lay down a light railway and use trucks. The line of boys is the only practicable way if the stuff has to be carried up a slope to discharge, or taken over irregular paths out of the work, as is often the case. The railway needs much time for rearranging different points of collection and discharge; and must be in duplicate, or else the work will be at a stand-still during rearrangement. A boy will carry 20 to 30 lbs. in a load, about 20 journeys an hour for 100 yards discharge, thus moving about 2 tons a day. So the cost is about a piastre a cubic metre for shifting 100 yards.

But far the more economical and rapid work is that of turning over whenever practicable. If a site has not been often rebuilt upon, the way is to start by a long clearance at one edge; and then a line of men steadily cut from one side of the trench and throw back on the other (Fig.24), so that the trench moves across the whole site, and every pound of earth is turned over. Each man needs a frontage of between 4 and 6 metres in width; and the trench, if open along, should have a clear bottom of at least 2 metres, from back to front of the work. More usually it is worked in compartments, each man clearing about 4 metres square, and throwing into his previous hole; each hole is then gauged when empty and the pay assessed. If a town is cleared (Fig.23), then it is done chamber by chamber, each being emptied over the wall into the previous chamber. The corners of the chambers can just be left visible for making a plan afterwards. A great advantage of this way is that the ground is finally left covered, so there is no great waste heap, and the walls are all covered over again to save them from future destruction.

Where a deep hollow has to be cleared out it is a wasteful plan to let the boys walk out with the basket of earth, as they have to raise the body, which is about four times the weight of their load. So soon as the rise is as steep as one in four, it is best to form a fixed chain of boys (Fig.27), each standing in a permanent place, and handing the baskets up from one to another. About 5 feet apart horizontally is as far as is useful; or in case of steep work (as out of pits) the vertical lift may be 3 or 4 feet (Fig.1). A sufficient number of collectors at the bottom and throwers at the top are of course needed to keep the chain in full work (Fig.28). A well-proportioned gang should not have any accumulations along it, and must be quietly watched from time to time to see that all parts work equally. If the baskets of earth lag at any point and accumulate, the boys before the point must be thinned, and those beyond it increased. A favourite plan of the boys is to let a basket lie unshifted and then stand upon it, as a full basket of earth gives a pleasant footing, and there is one less to keep moving. In this way most of the baskets can be quietly suppressed and yet every one remains as busy as they can be with the short stock of baskets that remain. All such misuse of baskets must be stopped at once; but old burst baskets may be used thus with advantage.

This system of lifting is also used in a surprising way for vertical tomb pits. An Egyptian man will stand all day with his feet on opposite sides of a pit in foot-holes, and stoop down to take a full basket from a man below at the level of his feet; then raise himself, and lift the basket up at arms’ length above his head, thus lifting it 6 or 7 feet. Three men will thus empty out a pit to 20 feet deep; but such men are usually old tomb-robbers, and must be employed with circumspection. More usually ropes are used, one tied to each handle of a basket, and pulled up by a pair of men. The earth is best left in the carrying basket, which is laid in the roped basket at the bottom, and taken out of it at the surface. If the pit is rotten and wide at the top, the basket has to be swung across the top two or three times, until on letting the ropes loose it flies out 10 or 20 feet to the side of the pit, where it is caught by the emptying boy. Clever rope-men will let a basket fly so as to catch on the top of the dump heap and turn over, so that it only needs clearing loose to let it go back again. The ropes need careful watching; the men love to tie knots in them, to grip by, whereby they wear through at the knots and drop to pieces; also the ropes are dragged on the edge of the pit, so as to serve as a friction-clutch when changing hands, thus wearing the rope out in two days instead of two months; the sides of the pit should be looked at to see if there is any sawing by the rope, and if so, the men must be stopped. They also cut off pieces if the ropes are long; and it is best to have all ropes in standard lengths of 8 metres, these when doubled thrice over down to 1 metre length are quickly tested for length, and then hanked in the middle to put by. Lastly, if not regularly delivered into store every night, the ropes are not returned when a pit is finished; and then they vanish, and a fresh pair is asked for when the next pit goes deep.

Another favourite misuse of ropes is to lash them round blocks of stone which have to be dragged, and thus cut the rope into scraps by wearing on the ground. Ropes can generally be put round the sides of a stone, and kept in place by some old scraps passing beneath.

One of the most careful kinds of work, to which only good men can be trained, is that of tracing out unbaked brick walls buried in rubbish. The surrounding earth is derived from the crumbling and washing down of the earthen wall, and therefore it is indistinguishable from the average of the bricks themselves. Hence, if the bricks are uniform in colour, and the mud mortar is like them, the building and its dÉbris are all alike. The best way to examine brickwork is by scraping a face of the wall, and then peeling it quite clean with a dinner-knife; such a clean smooth surface seen in shadow will show whatever can possibly be made out of the differences of colour and texture. Vertical joints are worth far more than horizontal, as often fallen bricks may lie as if built together. If possible the joints should be observed by differences of colour, and the bricks measured for comparison with others; as the sizes vary from 7 inches to 2 feet in length, and but seldom range over half an inch in any one building period, the size will go a long way in showing a connection of age. If the bricks cannot be distinguished even after leaving the face to dry for some days, the earth should be searched by pecking with a trowel or knife to see if there is dirt in it: only in late times are pottery chips found usually in bricks, and charcoal scraps are very rare, hence pottery and charcoal almost prove the earth to be mere wash and rubbish. The clearing back of dirty earth to a vertical face of clean clay is a satisfactory evidence of a wall. But sometimes the filling is so clean that there is no difference between it and the wall. Then the relative hardness will often serve to distinguish one from the other; and this is a main means of discrimination by the workmen, who will often tell a wall entirely by the touch under the pick. Failing all these tests, and the strata of dirt beds, the film of stucco on the wall face will sometimes show up, but may leave a doubt as to which side is the wall. In the last resource the stuff should be searched with a magnifier to see the hollows left by decomposed straw dust: in kneaded brick these hollows lie in every direction; in blown dust and wash they lie nearly all horizontal. It is often needful to spend half-an-hour testing and tracing out the line of a wall, fixing the face and the top and base of it; and such work may give the only evidence of a temple or important building.