342 History of Mechanical Inventions and 



set^ — the cutter is to be connected with the latter, and turned swiftly 

 round by it, exactly in the manner now done in these engines. 



Now, supposing the length of that part of the epicycloidal curve, which 

 is required for the tooth of the wheel, to be one-twenty-fourth of an inch, 

 which is nearly the size for the smallest sized chronometers, or those worn 

 in the pocket, than if the pentagraph diminished twenty times, the same 

 portion or the curve next the chocks, which is to be cut into the brass 

 frame, would be about an inch in length, which size is sufficient for it to 

 be constructed mechanically with sufficient accuracy, as any trivial errors 

 in that mechanical construction, when decreased twenty times, would be- 

 come insensible. 



The best kind of pentagraph, I should think, would be the eidograph in- 

 vented by Professor Wallace, and which is made by Bate, in the Poultry, 

 London, as it is found to be peculiarly accurate, and to possess several other 

 advantages, and for the suggestion of employing this pentagraph I am in- 

 debted to P. Barlow, Esq., of the Royal Military Academy. It ought to be 

 made of steel and hardened, to prevent it bending by the resistance opposed 

 to it, which, however, is not great, as the motion of the cutter is very swift. 



There are several little things to be attended to in putting this method 

 in practice, all of which would immediately strike the practical mechanic, 

 but some of which I shall mention. The mandrill would have to be length- 

 ened a little, to enable its top, with the wheel on it, to stand above this 

 second frame. This may be done by unscrewing the part which holds the 

 wheel out of the mandrill, after the first cut is made in it by the old cut- 

 ting-frame, and then screwing on an additional piece to the mandrill, and 

 again screwing it, the piece holding the wheel on the top of this additional 

 piece, or the second frame might have shoulders standing up above its upper 

 surface, at the end next the chocks b, b, to receive the hollow steel cones, in- 

 to which the inner conical ends of the screws/,^ come, which would hav- 

 the same effect. The approaching and receding motion of the pentagraph 

 must be exactly horizontal, or more properly speaking, it must be in a plane 

 parallel to that of the upper surface of the dividing plate, — the middle of 

 that part of the circumference of the cutter which is next the wheel, must 

 be the part which gives the cut, — when all is accurately set as to depth, &c. 

 every time the blind man's guide turns round the requisite portion of the 

 dividing plate, the tracing end of the pentagraph will have to be moved by 

 hand through the deeeply cut curve at the back part of the frame, the same 

 under similar circumstances, as the cutter frame now used, has to be 

 brought down by hand, to cut through the circumference of the wheel. 



There is a method a little different from the above, by which the teeth 

 of wheels might be curved epicycloidally, but I think it would be found 

 more difficult to use it with accuracy. I will, however, state it, and should 

 any public-spirited chronometer maker be inclined to this essential im- 

 provement in his machines, it will be for him to determine on this, and 

 many parts in the practical applications of the plans I give, which he will 

 be much more qualified to do than I am. 



Jn order to explain this, Fig. A, Plate I, shows the axis 1 1, of the cutter as 



11 



