Jg2 frECREi jlock; 



Detcrrptujh of ithrftugh the openTpac^s of evfery one of the wheels. Pig. 2. 



new I6ck of reprefenls a contrate wheel, having its irregular porlions 

 A, fi, C, D, &c. flanding up above its plane. Thefe por- 

 lions are parts of a circle equal to that denoted by the broken 

 ftiaded parts in Fig. J. The contrate wheel is to be placed 

 in Fig. I. with its face turned down ; and being there fcrewed 

 with its center to the central handle, it ferves to open and 

 flnit the bolt, which it can only do when the four wheels are 

 in fuch a filuation as to allow the circular edge-parts of Fig. 2. 

 io pafs clear through their notches. If any one or more of 

 thofe wheels be turned fo as not to correfpond with its num- 

 ber, it wil! be impoffible to turn the handle, becaufe every 

 attempt to do fo will caufe one of the parts of Fig. 2. to ftop 

 in one of the notches of the wheels through which it cannot 

 ]pafs. The method of opening the lock will therefore confift 

 ill fetting each wheel to its known number. 



As the proper fitualion of each wheel is only one out often, 

 ft is nine to one again ft any operator upon this lock, that he 

 ii)all not' fet th^ firft wheel right, fuppofing all the others in 

 their due pofitions ; but it is true that he may try all round, 

 and will coihe to the right place at laft. If two only of the 

 wheels were deranged, it would be eighty-one to one that he 

 fliouW not fet them both right ; and he wouSd be deprived of 

 any trial round a tingle wheel, becaufe the other wheel would 

 always hold againft him, and prevent his knowing when the 

 6pert rtotch of the wheel under trial prefenled itfelf. Three 

 wheels deranged would make the odds 729 to one, and the 

 four would make the odds 6561. In the plate the combina- 

 tions are faid to be ten thoufand, from an overfight in taking 

 the ratio of ten to one infiead of nine to one. But this is a 

 matter of no confequence as to the principles of the lock, be- 

 caufe the number of teeth or number of wheels are capable of 

 variation. If a fifth wheel were added to this lock, the odds 

 would amount to 590^9. 



As the quantity cut from Fi^. 2. is not more than was ne- 

 cefFary for the clear rotation of the wheels when the lock is 

 ihut, this piece, when in every other pofition, prevents the 

 other wheels from beinx turned at all. 



Leiter 



