TRACT 36.

THE WHIRLING-MACHINE.

169

when each hemisphere was taken off, to find the resistanceto the arm, and to obviate any objections, for the same rea-son as Mr. Robins had done before; the arm was fixed on,with the brass wire, strained from the top of the axis to theend of the arm, to sustain it in the horizontal position.

May 1 , 1736.

11. Began to take the necessary dimensions of the parts ofthe machine. Measured the diameter of the axis two differentways, and each way several times over, taking a mean ofthem all. One way was bv means of a sliding shot-gage,■which measured diameters very nicely to hundredths of aninch, by taking the diameter between two parallel arms,made to slide along a stem, divided into inches and lOths,and these into lOths again, or the whole into lOOths ofinches, by means of a nonius. In this way I took, severaltimes over, and always the same, the diameter of the barecylinder, without the silk thread wrapped, and then again thediameter including the silk thread on both sides, which is 2diameters of the thread more than the bare cylinder; viz,Diameter of the cylinder alone .... 2*06 inc.

Ditto and double diam. of the thread . . 2-11

Theref. diam. of cyl. and 1 diam. of thread 2*085And radius to middle of the thread . . 1*042

The other way was, by making the machine turn rtJundexactly a certain number of times, and measuring the spacedescended by a weight at the end of the thread, as it un-wound from the cylinder ; then the space so descended be-ing divided by the number of turns, gave the length of oneround or circumference of the cylinder and thread ; and thisagain divided by 3*1416, gave the diameter of the cylinderand thread included. This operation being often repeated,and always nearly the same, the medium of all gave for thediam. of cyl. and thread 2 092, therefore the radius of thesame is 1*046, which is only '004 part of an inch more thanthat by the former method ; and the medium of the two isV044,