SAND TRAPS 145 be derived from a completely closed whitewater system are obvious, and, in fact, all mills have always strived to achieve this end, which is so intimately bound up with the very important saving of fresh water and the elimination of effluent which may have to flow eventually into rivers. If steam has to be used to heat the stuff for heavy or wet beaten papers, the high-level wrater box is where it will give the most economical and efficient resuhs. The pipes of the water system, and indeed all pipes conveying back water and stuff, must be copper with brass or copper bolts and nuts throughout. The back-water pipes should be taken apart and thoroughly scoured out fre- quently. A tight bundle of strips of old machine wire pulled through by a good strong rope makes a very efficient scrubber. The water system of a machine should never be allowed to get dry. If this happens during a shut down, or over the week-end, a quantity of scale or slime is sure to break away when the water is put on, and may continue to come through and get into the paper for hours. The pipes should be left full, with a trickle coming through the water gate and a little overflowing. It will be found, in many mills, that the pipes are allowed to go uncleaned for months, in order to save the few hours this takes, or the few shillings paid for overtime to the fitters who take down and replace the pipes. This is the worst possible mistake, for a machine cannot produce clean paper if the water system is neglected. From now onwards we must deal with the stuff and water mixed in paper- making proportion, which may be from i to 2 parts of stuff to 100 parts of wTater. The Sand Traps.—These are the wooden shutes through which the stuff and water flow, leaving in their passage all heavy foreign matter, such as sand, specks of metal, etc., and any substances heavier than fibre. A difference of opinion exists as to whether deep, narrow shutes, or wide, shallow shutes are most efficient. It is contended by some that deep, slow- moving stuff gives the best chance for, say, a piece of metal to fall to the.bottom. Others say that the stream must be shallow and ripply to break up the clusters of fibres and allow the piece of metal to drop. The truth may be with both views, but there seems to be no reason why shutes should not be made with a deep channel in one part and wide and shallow in the other. There ought to be several depressions in the bottom of the shutes at intervals about 3 or 4 inches deep, in which dirt, metal, etc., may be trapped, otherwise it will gradually travel to the end of the shutes and get into the strainers. . It is a very good plan to have the bottoms covered with strips of old wet felt with a good nap and of a handy length to facilitate washing. These are held down by strips or bars of lead. Two sets of strips should be available,