concurr'd with contrary motions, A with twelve parts of motion, and B with ſix, then if A receded with 2, B receded with 8; to wit, with a deduction of 14 parts of motion on each ſide. For from the motion of A ſubducting twelve parts, nothing will remain; but ſubducting 2 parts more, a motion will be generated of 2 parts towards the contrary way; and ſo, from the motion of the body B of 6 parts, ſubducting 14 parts, a motion is generated of 8 parts towards the contrary way. But if the bodies were made both to move towards the ſame way, A, the ſwifter, with 14 parts of motion, B, the ſlower, with 5, and after reflexion A went on with 5, B likewiſe went on with 14 parts; 9 parts being tranſferred from A to B. And ſo in other caſes. By the congreſs and colliſion of bodies, the quantity of motion, collected from the ſum of the motions directed towards the ſame way, or from the difference of thoſe that were directed towards contrary ways, was never changed. For the error of an inch or two in meaſures may be eaſily aſcribed to the difficulty of executing everything with accuracy. It was not eaſy to let go the two pendulums ſo exactly together that the bodies ſhould impinge one upon the other in the lowermoſt place AB; nor to mark the places s and k, to which the bodies aſcended after congreſs. Nay, and ſome errors, too, might have happened from the unequal denſity of the parts of the pendulous bodies themſelves, and from the irregularity of the texture proceeding from other cauſes.
But to prevent an objection that may perhaps be alledged againſt the rule, for the proof of which this experiment was made, as if this rule did ſuppoſe that the bodies were either abſolutely hard, or at leaſt perfectly elaſtic; whereas no ſuch bodies are to be found in na-
