Instead of looping the rod around and back through the rolls he put several sets of rolls each in front of the other with every other pair vertical. This alternate horizontal and vertical roll arrangement was necessary because the reduction of the billet or rod in any pass can be only in a direction perpendicular to the axis of the rolls. The roller and catcher had given it a quarter-turn twist each time they started it into the two-high rolls. Bedson’s successive pairs of rolls were set close together, each pair being speeded enough that it took the rod exactly as fast as the preceding pair delivered it. The billet or rod, therefore, traveled through them in a straight line. This continuous process was, of course, of great advantage in that considerable speed could be attained and there was not the rapid cooling nor the opportunity for loss by scaling from exposure to the air which occurred with the long loops of the Belgian Mill. It was a great advance, but the speed was yet held down by the finishing pass and the inability of the reel to coil the rod fast enough. The greatest development came through the inventive genius of two men, Charles Morgan and George Garrett, who developed the two separate types of mill which have made the rod rolling industry what it is to-day. The work of both was done in this country. The Morgan Continuous Mill Morgan’s also was a continuous process. The billet was put in at one end of a new type of heating furnace which Morgan devised, and was gradually pushed along to the other end. From this second or outgoing end, the long, narrow, white-hot billet went through several pairs of two-high rolls set close together, each successive pair having smaller grooves than the one preceding it, so that, after traversing these several pairs of rolls, the rods emerged from the last pair finished, having traveled in a straight line through them. Hand reeling was much too slow for Morgan who invented two different types of high-speed automatic reel which, in the highly speeded mills of to-day, receive and coil wire coming from the finishing passes at speeds of over one-half mile per minute. It is stated that the billets and rods therefrom traverse the rolls so fast and the pressure is applied so quickly and so hard that the rods emerging from the finishing passes are hotter than were the billets when they went into the rolls. [Illustration: THE BILLET TRAVERSES THE MORGAN CONTINUOUS MILL AT HIGH SPEED EMERGING FROM THE LAST ROLL AS FINISHED ROD ] As was explained, no reduction in thickness is brought about by the sides of the grooves in the rolls. Therefore, a bar or rod must be turned after each pass unless Bedson’s alternating horizontal and vertical rolls are used. This turning Morgan did in spiral tubes inserted between the successive sets of rolls, all of which were horizontal. These tubes operate as does the “rifling” in a gun barrel, in turning the rod.