422 Mr DENISON, ON SOME RECENT 



that the application of oil made very little difference in the arc; and small as it is, it 

 might be made still less by using a double scape-wheel (as suggested in my Rudimen- 

 tary Treatise, for the common dead escapement), with long teeth for the locking and 

 short ones for the impulse ; in fact, one has since been made in this way in Paris by a person 

 to whom a drawing of the escapement was given. 



It is worth remarking too, that I found the clock always gained with an increase of 

 arc, even though it was produced by increasing the clock -w ei gh t : thus affording another 

 proof of the uselessness of any such accelerating contrivances as I before mentioned. 

 And as a small recoil had been given to the dead pallets, it would perhaps be possible 

 so to adjust the amount of this recoil, that the pendulum should neither gain nor lose 

 for any probable amount of variation of the arc. And at any rate, as the force of the 

 scape-wheel may be kept almost constant by a remontoire in the train, and any variation 

 of the dead friction may be made quite insensible by the 'duplex' arrangement just now 

 referred to, there can be little doubt that this escapement might be made more constant 

 in its action than any other in which the pendulum receives its impulse from a scape-wheel. 

 It requires however more delicacy in construction than the single-pin escapement, on 

 account of the small depth of the intersection of the circles described by the points of 

 the scape-wheel teeth and the corners of the vertical pallets ; and it probably could not 

 be applied to a watch at all. 



The crutch, or arm which carries the pallets, must be about 25 times as long as the 

 legs of the scape- wheel, in order to let the escape take place at 1° (there will always 

 be a little lost in the drop) ; and therefore the legs cannot well be more than £ inch long 

 in an astronomical clock. In a turret-clock they may be two or three times this size. 

 The scape-wheel which drove the pendulum of the Westminster clock (weighing above 

 6 cwt. and 14^ ft. long), was only half an inch in radius, and 1 of an ounce in weight; 

 and though the (temporary) pallets were only of iron case-hardened, the teeth had not 

 made a mark upon them after it had been going for four months, when it was taken 

 off to be replaced by another escapement which I have yet to describe. 



For after it had been going for some time, I was consulted about a clock for the Cathe- 

 dral at Fredericton, where the cold is sometimes as great as - 40° ; and some apprehen- 

 sion was expressed as to the possibility of making any ordinary escapement to go at that 

 temperature ; I think not without reason, as even chronometers will stop at a less degree 

 of cold than that, and a watch with a lever escapement (which is the common dead 

 escapement without the dead friction) will stop at a temperature far above it. I have no 

 doubt the three-legged escapement would keep a clock going through any cold which does 

 not actually freeze the oil in the pivot-holes. But as even oil which will not freeze is ren- 

 dered much less fluid by extreme cold, the force of the escapement would probably vary 

 so much as to make the pendulum swing a good deal further in summer than in winter, 

 and thereby affect the rate. 1 therefore resumed the attempt, which I had long abandoned, 

 to invent a gravity or remontoire escapement, capable of satisfying all the requisite me- 

 chanical and mathematical conditions, and sufficiently easy to make. And I found that 

 it could be done by turning this three-legged impulse escapement into a gravity escape- 



