time in changing the reels and cutting the paper 

 loose from them, two reels were hung at opposite ends 

 of center-pivoted arms. Mr. Gilpin had attached to 

 each reel a counter with dial showing the number of 

 revolutions or sheets wound. 



When a reel had on it the number of sheets required, 

 the attendant would swing it over without stopping 

 the paper being formed, and as soon as the empty 

 reel had taken the place of the full one he would, 

 with a stick, break down the wet paper pressing the 

 end between the slats of the empty reel, and the 

 winding would continue on it. 



Mr. Gilpin complained of the necessity of having 

 so many different sizes of reels, or being obliged to 

 change slats to suit the different sizes of paper ordered. 



I asked why not make the reels to expand and 

 contract to adjust for the different sizes? 



He turned to me and in his quick, impulsive manner, 

 said: "Boy, that is easier said than done. I have often 

 thought of it but I cannot see any way of accomplish- 

 ing it." 



His sudden explosion almost took my breath from 

 me; I felt as a boy, I had been too hasty in putting in 

 my oar. I stammered 1 : "If you will come to our 

 shops, and see grandfather's old sieve hoop rounders 

 or stretchers, they may give you an idea." 



These old stretchers were nothing more than a plate 

 scroll screw, that, on turning, thrust out a number of 

 arms against the inside of a sieve hoop holding it to 

 a perfect circle, while the wire cloth was being tacked 

 on — the same in principle as the present universal 

 lathe chuck. It was not long until Mr. Gilpin came 

 to see the stretchers, and was so well pleased that he 

 undertook to have a couple of six-arm expansive reels 

 made. 



At that time there was no face lathe in Philadelphia 

 with arrangement for cutting the scroll screw. He 

 had recourse to Isaiah Lukens, who wound a wrought- 

 iron scroll thread, riveted and brazed it on to a 

 wrought plate in the same manner grandfather had 

 made his original stretchers. Mr. Gilpin was very 

 proud of these reels, so much so that he had a grad- 

 uated scale with vernier attached, that enabled him 

 to at once adjust for any length of paper; this, with 

 his counters and their registers, he considered a great 

 advance, but they were to be superseded by steam 

 dryers, calenders, slitters, and cutters to deliver the 

 finished sheets direct from the machine. 



Sometime previous to the visit of Mr. Gilpin's 

 mill a very simply constructed cylinder was brought 

 to be covered with wire cloth. I cannot recall the 



name of the Frenchman who brought it nor the 

 location of his mill except that it was in New Jersey, 

 and that it was a hand mill making wrapping and 

 sugar loaf paper. It was a small affair, not over 

 a feet 6 inches long and about 2 feet in diameter. 

 It was constructed on the principle of a squirrel cage 

 wheel. Six slender-arm brass wheels on an iron 

 shaft that was covered with a copper tube formed 

 the framework of the cylinder. The end wheels had 

 rims about i% inches deep by %-inch thick, and at 

 their smallest diameter a projecting rim or collar 

 about i inch, not over %-inch thick at its outer edge. 

 The four center supporting wheels equally spaced 

 on the shaft had round rims of about %-inch diameter. 

 Holes were drilled close to the outer edge of the end 

 wheels about I inch apart of a size to take in a No. 6 

 brass wire through these holes. Brass wire rods of 

 that size were run from end to end of the cvlinder, 

 resting on and being supported by the round rims 

 of the intermediate wheels. This was the condition 

 of the cylinder when it was brought to the shop of 

 N. & D. Sellers to be finished and faced with wire 

 cloth by its inventor and maker, and I recollect it 

 as a very creditable piece of work. The rims and 

 collars of the end wheels were accurately turned, the 

 arms and hubs finished, as were also those of the 

 supporting wheels whose round rims were turned as 

 much as they could be beyond the arms, the portions 

 between the arms file and scraper finish. The man 

 had worked out in his own mind exactly what he 

 wanted done. In the first place the longitudinal 

 rods must be secured to the round rims of the sup- 

 porting wheels so as to keep them straight and equal 

 distance apart; this he proposed doing by tieing to 

 the rim with soft copper wire. It was done by the 

 kind of loop sewing as wire screens were then made. 

 This being done the cylinder was wrapped from end 

 to end with No. 18 hard brass wire laid so as to leave 

 a space of )\ inch between each round. This was 

 secured to the longitudinal rods in the same manner 

 of loop sewing as they were secured to the rims of 

 the supporting wheels, then covered as the Gilpin 

 cylinder was by a wove brass wire backing of 14 

 meshes to the inch, over which the fine wire face 

 was placed. 



The explanation given of the object of the projec- 

 tion rims or collars on the ends of the cylinder, was 

 that the cylinder was to be sunk, to say, % or 3 ^ of its 

 diameter in a vat of dilute pulp, these collar projec- 

 tions coming into close contact with the sides of the 

 vat. By tacking a strip of sheep skin with the wool on 



99 



