This is usually achieved by making a very thin crepe and rolling together two lengths when wet. The resulting crepe may be slightly thicker than ordinary, and the method employed may be usually detected by the appearance of the edges unless these are trimmed. GREENISH AND TACKY STREAKS.--Occasionally one meets cases in which pale crepe exhibits streaks varying in colour from a decided green to an almost black in which the greenish tinge is scarcely perceptible. Experience indicates that these streaks are much more dangerous than they appear superficially, inasmuch as they contain traces of brass from the "liners" of the bearings. The presence of the copper in brass is responsible for a gradual disintegration of the rubber, commonly recognised as "tackiness." In fact, copper may be said to be a "poison" to rubber, and every effort should be made to avoid possible sources of contamination. The effect may be proved easily and perceptibly by fastening together several pieces of crepe rubber by means of a brass "paper-fastener." In course of time a salt of copper, green in colour, will be formed, and it will be found that the portions of rubber in contact with the fastener have "perished" and become tacky. This contamination of crepe rubber may take place in two ways: (1) By the ejection of actual particles of brass from the bearings of machines, due to eccentric grinding of the "standards" of the rolls upon the brass "bushes." These particles are carried by exuded oil or grease into trays, and thence to the rubber. (2) By the action of an acid lubricant upon the brass, with the formation of a metallic soap which has a decided green colour, unless obscured by the dark colour of the oil or grease. It is transferred to the crepe rubber in the manner indicated above. The inevitable effect, apart from the superficial defect, is incipient tackiness. The extent to which this may develop will depend upon the amount of the copper compound present, but it should be remembered that an exceedingly small trace is capable of causing a disproportionately large amount of damage. This effect is further magnified if the "tacky" piece of rubber is packed in close contact with previously unaffected rubber. When the defect is discovered, the affected portions should be cut out, and the cuttings should be burned. To mix them with the lowest scrap grades, as may be done thoughtlessly, is only inviting further trouble. Besides the source of danger already indicated, it may be found, but far less frequently, that contamination may arise from the presence in the rubber of small pieces of the brass mesh which is generally used for straining latex.