DIRTY EDGES.--It seems to be almost impossible to keep old machines clean, and it is equally difficult to keep the edges of crepe free from oil and dirt. Usually these dark edges are to be found on crepe which is rather wide, and it will be noticed that where wide crepe is made, unless special precautions are taken, the edges of the rubber often pass under the edges of the hopper and so pick up dirt and oil. On most machines it is a great mistake to attempt the preparation of wide crepe; nothing but narrow crepe must be made. To obtain this it is necessary to decrease the width of the hopper placed above the rolls. This can easily be effected by blocks of heavy hard wood, cut to shape and fastened in position. Sometimes the dark edges of crepe are due to another cause. Rolls may be gradually worn in the middle, so that to obtain a good finish it becomes increasingly necessary to tighten up the screws which regulate the distance between the rolls. It thus happens that just at, and beyond, the edges of the rubber the rolls grind upon each other, and fine particles of iron and graphite are transferred to the rubber. In such a case it is evident that either the rolls must be "turned" or that a new pair of rolls must be substituted. IRON-STAINS.--One of the causes of iron-stain on rubber has been mentioned in the preceding paragraph. This particular kind of iron-stain must not be confounded with rust-stain, and gives a dark dirty colour. It results from the grinding together of the rolls, and is usually noticed in the finishing of fine pale crepe. For this operation it is necessary to screw up the rolls tightly, and it will be plain that, whenever the rolls are vacant of rubber, there is a tendency for them to grind upon each other, thus setting free fine particles of iron and graphite. In order to avoid this, one must be careful to see that between the working of each length of fine crepe the rolls should be occupied with another piece of rubber, which may be kept for the purpose. In some factories this trouble apparently does not exist, while in others the amount of wear on the rolls is surprisingly great, and the damage done to the rubber is excessive. The only way in which this difference can be accounted for is that there must be a great difference in the quality of the roll material. Some rolls seem to be excessively soft, and from these contamination by iron-stain is great. For this reason rolls are sometimes rejected, and there would appear to be an objection to any but chilled steel rolls for the final stage of finishing crepe rubbers.