so leaves a clean fractured surface all over. It must be remembered that bricks are often bent out of form by solid flow of the wall under great pressure, so that they may be distorted almost like a glacial deposit. In cleaning down the face of a wall it may often be traced by its hardness, but this is not a test to be left to workmen, or they may cut away at random; a very good plan is to let the man trench along a few inches outside of the face of the wall, and then cut down the remaining coat of rubbish oneself, to bare the face. Though pottery, stones, &c., often serve to show what is accumulated soil, yet they are found in brick sometimes, and must not be relied on entirely. The texture of the soil is important, as in accumulations all long bodies, bits of straw, &c., lie flat; whereas in brick they are mixed in all directions. Also washed-down earth almost always shows worm casts in it. Often a wall, if in low wet soil, will show out distinctly when the cut surface has dried, as cracks will form more readily along the joints. In many cases, however, all of these tests hardly serve to unravel the puzzle; especially where there are successive walls superposed, and only a small height of any one to examine. To trace out the position of ancient walls is, however, one of the first requisites in such work; not only do we recover the plan of the town and its buildings, but we are led thus to recognize what may be the most important sites for special excavation. One of the most difficult questions always is to know what may be safely thrown away. Most trivial things may be of value, as giving a clue to something else. Generally it is better to keep some examples of everything. No matter how broken the potsherds may be, keep one of each kind and form, replacing it by more complete examples as the work goes on. Thus the collection that is kept is always in process of weeding. It need hardly be said that every subject should be attended to; the excavator’s business is not to study his own speciality only, but to collect as much material as possible for the use of other students. To neglect the subjects that interest him less is not only a waste of his opportunities, but a waste of such archaeological material as may never be equalled again. History, inscriptions, tools, ornaments, pottery, technical works, weights, sources of imported stones, ethnology, botany, colours, and any other unexpected subject that may turn up, must all have a due share of attention. And keeping up the record of where everything has been found, and all the information that will afterwards be needed, about the objects and the discoveries, the measurements and details for publication, is a serious part of the work.