22 Experiinintal Rtfiills and Ol'j'ervatioits 



mcrs, a fcliemc upon wliich many thoufands of pounds have been expended in this coun- 

 try, it was apprehended that the more impure parts, whioli are alio the mod fluid, might 

 be preflcd out by the aiflion of cylinders, witli equal or perhaps more advantage than by 

 iliat of hammers ; at the fame time that the determinate figure of bars of any required 

 fizc might be given without Ikill in the operator. Experience neverthelefs has (hewn, 

 that the more fluid part is driven out much more efFedlually by the fudden a6lion of a 

 blow, than by the flower comprelTion of a cylinder, which allow 3 time for much of the 

 fluid matter to extend itfcif within tlie mafs. Various fimilar rfTe^ls prcfciit themfelves 

 when cylinders for printing are fubftituted inflcad of planes. Inllead of the a£lion of 

 dabbing, the colour is ufually applied by fimplc and gradual contaift, to much Icfs eflecl ; 

 and the impreiTion, thougli not cflentially ditVerent from that of the block, is performed 

 by a gradual aCtion, which aiTords time for the cloth or paper to fold itfelf in a minute 

 degree into the cavities of the fculpture. Hence it is found that the length of paper or 

 cloth printed from a cylinder by a definite number of revolutions, will be greater or Icfs 

 than another piece manufa£\ured preclfely in the fame way, but wi:h a lefs or greater 

 degree of preffiire. In a block this defecfl is much lefs, not only from the confiderable hold 

 it takes upon the furface of the material, but alfo becaufe the error is rectified at every 

 fucceflive application. One of the chief difiicuities of cylinder printing confifts, there- 

 fore, in the difficulty of laying one colour after another ; and this would continue to be fo 

 even if the materials were not fufceptible of change, the contrary to which is the fail. 

 There are two projc£ls for obviating this. The one confifts in confining the whole piece 

 to a long table, or to the circumference of a large cylinder, and caufiiig the printing cylin- 

 der to move, not by the fuccefTive appofition of its carved furface, but of a bearing face 

 regulated by a toothed wheel. The other method confifts in the ufc of a frame to confine 

 two or more cylinders, each provided with its own toothed wheel, and revolving againft 

 a large clotlied cylinder provided with a fuitable wheel to drive the others. Tlie piece is 

 caufed to pafs between the large cylinder and the others, in order to receive the impreftion 

 With regard to the firft of thefe methods, it does not appear eafy to confine paper, and ftij 

 lefs cloth, in fuch a manner that its parts may continue without (hift or wrinkle during the 

 a£lion of a cylinder, which not being allowed to roll without the check of a wheel, mud 

 draw the furface either the one way or the other. The difficulty of confinement will be 

 very much increafed, by the indifpenfable requifite that the paper fliould be afterwards hung 

 up to dry, and the callico be carried to the dye-houfe and the bleach-field, between the 

 fucceflTive impreffions, by which means the dimenfions of both will be greatly altered. In 

 the fecond method, it is obfcrvable that no colours can be printed but fuch as fall clear 

 of each other. In this way, moreover, the gathering a£lion of the cylinders may prove 

 very mifchievous. For, if we fuppofe the paper or cloth to pafs between the great cylinder 

 and the firft printing roller by an aclion of the latter which tends to make it flip forward 

 on the face of the great cylinder, and that when it arrives at the fecond printing roller it 

 there experiences anaflion of a contrary nature, the confequencc will be, that the material 

 will become flack between the two rollers, and the fittings will be falfe. Not to dv/eil on 

 that experience which brings forward this obftacle among others, its great probability may be 

 deduced from the allowable fuppofition that the circumference of the fiift printing cylinder 

 ihould be one thoufandth part of an inch too large, and that of the fecond the fame quantity 

 4 too 



