J50 **^« GREGORY Oy HORSE POWERS. 



Defcription of a bar of the harnefs of a fecond horfe, equal in point of ftrength 

 TamuTfor ^^' ^° ^^® ^'^'■"^^''- Then if the two horfes, thus attached to the 

 mcafuring the ears E and F, be made to pafs over the walk in the fame 

 fe-aftioninmiUs.^Ireaion, following each other conftantly at the diftance of a 

 femi-circumference; while that which draws at the ear E 

 overcomes the whole predbre and refi fiance oppofed by the 

 work, the other which draws at F by the cord over the pulley 

 p, will raife the weight w of the fleelyard; which therefore, 

 by being moved to and fro upon the arm/?, may be brought 

 to exhibit an exadl courilerpoife, or meafure of the exertion 

 and power of the horfc. And in order to ent'ure the |jreateft 

 degree of accuracy in this refped, the motion of the two 

 animals, and the polltion of the weight w, (liould be fo ad- 

 jured, that the fame weight fliould be (hewn by the graduations 

 both of the fpring and of the lever fteelyard. The fliaking of 

 the machinery w\\\ in iome meafure diflurb the effed; but an 

 ingenious manager of the exp^pments will find means of 

 *; checking this : and as to the centrifugal force to which the 

 weight w is expofed, it will never be of any material con- 

 fequence in any of the flov/ motions which will be produced 

 by this kind of worl^. 



Each experiment (hould occupy the fpape of a fair day's 

 work for the horfes: for the conclufions deduced from 

 ihorter and irregular efforts are always errqneous in excefs, 

 ^nd fliould be guarded againll. The rate at which the ^nim^ls 

 jnove may readily be afcertained froni the known circumference 

 iDf the walk, and the number of round#^1lhey are obferved to 

 make in 10 or 15 rainptes. Thu?;, by continuing the experi- 

 pients day after day, varying the velocity of the motion in 

 fome cafes, and the radius of walk in others, fuch a feries of 

 refults might at length be obtained, as would in a great 

 meafure remove the obfcurity and dopbt in which this bulinef^ 

 is at prefent enveloped. It is fcarcely neceffary to fuggefi 

 the propriety of making a few experiments with a view of 

 * (determining how far a load upon the back of each draught 

 jiorfe, would affift hrwHn his labour. No«;,cgn it b.e requifitQ 

 to point out in what way, by mean^ of fuch fieelyard's properly. 

 applied to waggons, &c. iipgji tolerably fmooth rpads, and 

 two horfps marching abreaO, (otiiJ|#rawing the load, the other 

 raifing the weight,) experiment*, miglit be inftituted tq 

 ?ifcertain the :i:a^nitude of tiie efFcrts cf horfcs when drawirig 



*■* 



