XXXII .
An Investigation of the Principles of progressive and rotatory Motion .
By the Rev. S. Vince , A. M. of Sidney College , Cambridge .
Communicated by George Atwood , A. M. F. R. S. Read June 15 , 1780 .
THE communication of motion from impact is well known to constitute a considerable part of that branch of natural philosophy called mechanics ; and as all our enquiries therein are directed , either to assist us in those operations which add to the conveniences of life , or to explain , for the satisfaction of the mind , those changes which we daily see arise from the effects of bodies on each other , it might naturally have been expected that the attention of philosophers would have been engaged , first in the investigation of such cases as most frequently occur from the accidental action of one body on another , before they had proceeded to others less obvious .
A little consideration will convince any one how seldom it happens , in the collision of two bodies , that their centers of gravity and point of contact lie in the line of direction of the striking body , yet few writers on mechanics have extended their enquiries any further than this simple case .
It must however be acknowledged , that the action of bodies on each other , in directions not passing through their center of gravity , affords a subject at least curious in speculation ; for my own part , I have little doubt but that it might be rendered extremely useful to the practical mechanic .
I. BERNOULLI was the first who published any thing on this subject .
He found the point about which a body at rest would begin to revolve when struck by another body , observing however that D. BERNOULLI had also discovered the same : he has also mentioned the curve described by that point in the progressive motion of the body , and has directed a method of enquiry by which the velocities of the bodies may be found after the stroke , which comprehends all he has done on the subject .
Two years afterwards D. BERNOULLI published a paper on progressive and rotatory motion , containing nothing more than what I. BERNOULLI had before given us , and , what is a little extraordinary , says in the introduction , de tali quidem percussione nibil adhuc , quantum scio , publici juris factum fuit ab iis , qui de motu corporum a percussione egerunt .
EULER has also investigated the velocities of the bodies after impact in a manner somewhat different , but has rendered it much more intricate by a fluxional calculus .
To any one , however , who attentively considers the subject , the theory must still appear to be extremely imperfect , as , independent of principles not more self-evident than the propositions they are intended to demonstrate , which both I. and D. BERNOULLI have assumed in their investigations , a great variety of other circumstances equally interesting naturally arise in an enquiry into this matter , circumstances absolutely necessary towards understanding the principles of the motion of the bodies after impact .
This induced me to consider the subject with some attention , and presuming that I have not been altogether unsuccessful in my endeavours to render the theory more perfect , I determined to lay the result of my enquiries before the Royal Society .
I thought it expedient , for the sake of perspicuity , to divide the whole into distinct Propositions ; and as the most simple cases are best understood , I have first considered the case of the action of a body on a lever having a corpuscle at each end : and I was the more induced to treat the subject in this manner , as most of the principles can be immediately applied to any number of corpuscles , in consequence of which the general investigations are rendered more easy and satisfactory .
PRO P. I. Let A and B be two indefinitely small bodies connected by a lever void of gravity , and suppose a force to act at any point D perpendicularly to the lever , to find the point about which the bodies will begin to revolve .
From the property of the lever , the effect of the force acting at D ( fig. 1 .
) on the body A is to the effect on B as BD : AD ; hence the ratio of the spaces Am , Bn , described by the bodies A and B in the first instant of their motion , will be as BD/ A : AD/ B ; join mn , and if necessary produce that line and AB to meet in C , which will manifestly be the point about which the bodies begin to revolve .
Hence from similar figures BC : AC : : AD/ B ( ~Bn ) : BD/ A ( ~Am ) : : A~ AD : B~ BD , or DC DB : AD + DC : : A~ AD : B~ BD , and consequently DC =A~ AD2 +B~ BDe/ B ~ BD A~ AD , and therefore D is the center of percussion or oscillation to the point of suspension C. Cor. 1 .
Hence , whatever be the magnitude of the stroke at D , the point C will remain the same .
Cor. 2 .
If the force acts at the center of gravity G , the bodies will have no circular motion , for in this case B~ BD A~ AD = 0 , and therefore DC becomes infinite .
Cor. 3 .
If the force acts at one of the bodies , the center of rotation C will coincide with the other body .
Cor. 4 .
If the lever had been in motion before the stroke , the point C , at the instant of the stroke , would not have been disturbed .
PRO P. II .
Let a given quantity of motion be communicated to the lever at D , to determine the velocity of the center of gravity G. The space Am , described by the body A in the first instant of motion , is as DB/ A : now CG = CD DG = CD AG + AD =Ax AD2 +Bx BD2/ B x BD Ax AD AG + AD =Bx BD x BG +Ax AD x AG/ B x BD Ax AD ; also CA = CD + DA =Ax AD2 +Bx BD2/ B x BD Ax AD + DA =Bx BD x AB/ B x BD Ax AD ; hence we have Bx BD x AB/ B x BD Ax AD ( AC ) : BD/ :A ( ~mA ) : : Bx BD x GB +Ax AD x AG/ B x BD Ax AD ( CG ) : Bx BD x GB +Ax AD x AG/ A xBx AB ~G~ the velocity of the center of gravity : hence if the motion be communicated at G , the velocity becomes as Bx GB2 +Ax AG2/ A xBx AB .
Let now the motion , which is supposed to be actually communicated to the rod at D , be equivalent to the motion of a body whose magnitude is G , and moving with a velocity v ; then if that motion be communicated at ~ , the velocity of the center of gravity is well known to be =Gx v/ A + B ; hence Bx BG2 +Ax AG2/ A xBx AB : Bx BD x BG +Ax AD x AG/ A xBx AB : : Gx v/ A + B : Gx v/ A +xBx BG x BD +Ax AD x AG/ B x BG2 +Ax AG = the velocity of the center of gravity , when the same motion is actually communicated to any point D. Now BD = BG + GD , and AD = AG GD ; hence Bx BG x BD +Ax AD x AG =Bx BG2 + Ax AG2 + GD xBx BG Ax AG = ( because Bx BG Ax AG = 0 ) Bx BG2 +Ax AG2 ; consequently the velocity becomes Gx v/ A + B ; and hence the center of gravity moves with the same velocity , wherever the motion is communicated .
PRO P. III .
Let a given elastic body P , moving with a given velocity , be supposed to strike the lever at the point D in a direction perpendicular to it ; to determine the velocity of the center of gravity G after the stroke .
Suppose first the body to be non-elastic , and let v be the velocity of the center of gravity after the stroke upon that supposition , and V the velocity of the striking body : then CG : CD : : v : vx CD/ CG = the velocity of the point D after the stroke , or of the body P ; for the same reason vx CA/ CG and vx CB/ CG equal the velocities of A and B respectively .
Now , because in revolving bodies , the momenta , arising from the magnitude of the bodies , their distance from the center of rotation and velocity conjointly , remain the same after the stroke as before , we shall have PxVx DC =vx CD2 x P/ CG +vx CA2 x A/ CG + vx CB2 x B/ CG , and therefore v=PxVx DC x CG/ P x DC2 +Ax AC2 +Bx BC2 = PxVx CG/ A +Bx CG +Px DC ; hence if P be supposed an elastic body , we shall have 2xPxVx CG/ A +Bx CG +Px DC for the velocity of the center of gravity after the stroke , in ipso motus initio .
PRO P. IV .
To determine the motion of the bodies after the first instant , or when they are left to move freely by themselves .
The writers on mechanics , from considering the equality of motion on each side the center of gravity , when a body revolves about that point , have inferred , that if a body had a projectile as well as a circular motion communicated to it , the center of gravity would continue to move in a right line , as that point would not be disturbed by the rotatory motion : yet , as , in the case we are now considering , the bodies begin to revolve about a different center , it may be proper to examine more accurately into this matter , and to show from what principle it is that the motion of the center of gravity is preserved in a right line .
Let a motion perpendicular to the rod be communicated to A ( fig. 2 .
) and then by Cor. 3 .
Prop. I. B will not be disturbed by such an action ; and A will in the first instant have a tendency to revolve about B as a center , and would actually describe the arc AH , if the body B were fixed : let the angle ABH be supposed infinitely small , and let GK be the arc , the center of gravity would have described , and draw the tangents AF , Gg to the arcs AH , Gg respectively .
Now , if A could have moved freely , it would ( because AF = AH ) have described AF in the same time the arc AH was described , upon supposition that B was fixed ; for the radius BA being perpendicular to the circular arc AH , the force of the lever could have no efficacy to accelerate or retard the motion of A in the arc AH , and therefore the velocity in that arc is the same as it would have been if it had moved freely in the tangent : hence HF is that space through which the centrifugal force of A would have carried that body , could it have moved freely ; but as A is connected to B by means of the lever , it is manifest that the same force which would have carried A from H to F in the direction of the lever , will , when it has both bodies to move , carry it over a space which is to FH as A : A+ B , or as Bg : BH , or as gK : FH ; hence that space , or the space through which the centrifugal force of A will draw the lever in the direction BH , is equal to Kg ; that is , the point K , which is the center of gravity of A and B , will be found at g , and consequently the center of gravity has preserved its motion uniform in the right line Gg , inasmuch as the centrifugal force , acting perpendicularly to the direction of the center of gravity , can neither accelerate or retard its motion .
In the same manner it may be proved , that the motion of the center of gravity is continued uniform in the same right line , whatever be the position of the lever .
Moreover , as the centrifugal force acts in the direction of the lever , it cannot alter its angular velocity , which will therefore remain as in ipso motus initio .
If now we suppose that to the force impressed upon A , two other equal accelerative forces be communicated to A and B at the same time , it is evident that no alteration can arise from the actions of the bodies on each other ; and the case will then be similar to the motion of the bodies , supposing a single force had been impressed at any point D. The like method of reasoning may be extended to any number of bodies .
The same thing may also easily be demonstrated in the following manner .
The centrifugal forces of A and B ( fig. 1 .
) are respectively Ax AC and Bx BC ; also the centrifugal force of the point G , considering it as having both bodies to move in the direction of the rod , is A+Bx GC , but from mechanics Ax AC +Bx BC =A+Bx GC : hence the centrifugal forces of the bodies A and B give the center of gravity a centrifugal force equivalent to its own centrifugal force , which , as the latter would cause that center to move in the tangent Gg , the lever not being fixed at C , it is manifest that the former will cause the center of gravity to continue its motion in the same direction .
That this motion of the lever , in a direction from the center C , is the only motion which is communicated to it from the effect of the bodies A and B is manifest from hence .
The bodies begin to revolve freely about the point C , and consequently if the point C had been fixed , the bodies would have moved on with a uniform angular velocity about C ; if therefore we suppose the lever not to be fixed at C , as the efficacy of the centrifugal force which acts in the direction of the lever is now suffered to take place , and no new external force is impressed on either of the bodies , it is manifest , that if in the former case the bodies had no efficacy to disturb the angular velocity of the lever , they cannot have any in the latter , consequently the angular velocity , and from what has been before proved , the uniform motion of the center of gravity in a right line , remain unaltered , after the commencement of the motion .
PROP .
V. In the time the bodies make one revolution , the center of gravity will move over a space equal to the circumference of a circle whose radius is CG ( fig. 1 .
) From the last Proposition , the angular velocity of the lever is continued uniform ; hence the time of a revolution is just the same as if the point C were fixed , and the bodies were to continue to revolve about that point as a center , in which case the center of gravity G , in the time of a revolution , would evidently describe the circumference of a circle whose radius is GC .
This therefore is the space the center of gravity describes in a right line when the bodies move freely , for from the last Proposition that center is carried uniformly forward with the same velocity .
Cor. 1 .
Hence if the magnitude of the force acting at D vary , the velocity of the center of gravity will vary in the same ratio as the angular velocity .
Cor. 2 .
Hence the point D may be found , where a force being applied , the bodies shall make one revolution , tion , whilst the center of gravity moves over any given space ( s ) : for let p= the periphery of a circle whose radius is unity , then p : 1 : : s : s/ p = the radius of a circle whose circumference is the space to be passed over in the time of a revolution , and which must therefore , by the Proposition , be equal to GC ; the point C therefore being determined , D may be easily found , for from mechanics CGxDG is given ; and from Cor. 3 .
Prop. I. when D comes to A , C will coincide with B , : CGxGD=AGxGB , and consequently DC=AGxGB/ CG .
PROP .
VI .
To determine the time of one revolution , supposing every thing given as in Prop. III .
The point D being given , we have from Cor. 2 .
to the last Proposition , CG=AGxGB/ DG ; put W equal the circumference of a circle whose radius is CG , and it appears from the last Proposition , that W is the space the center of gravity passes over in the time of one revolution ; hence , because from Prop. IV .
the center of gravity moves uniformly , we have by Prop. III .
2xVxPxCG/ A+BxCG+PxDC : 1~ : : W : Wx 2xVxPxCG/ A+BxCG+PxDC = the time of one revolution .
Cor. Hence the angular velocity being inversely as the time of a revolution , will vary as A+BxCG+PxDC/ VxPxCGxW .
PROP .
VII .
The point C , as the center of gravity moves forward , will describe the common cycloid .
From the description of the common cycloid it appears , that the center of the generating circle passes over a space equal to the circumference of that circle whilst it makes one revolution .
With the center G ( fig. 3 .
) and radius GC , describe the circle Cxy , and draw CR , GW perpendicular to ABC , and let the circle Cxy be supposed to revolve on the line CR ; then will the center G move over a space equal to the circumference of the circle Cxy whilst it makes one revolution , and the point C will describe the common cycloid : but from Prop. V. the point G will move over a space equal to the circumference of a circle whose radius is GC , whilst the bodies , and consequently GC , make one revolution ; and hence the point C will describe the same curve as before , that is , the common cycloid .
PROP .
VIII .
Let a motion be communicated to the lever obliquely , to determine the point about which the bodies begin to revolve .
Let FD ( fig.4 .
) represent the force communicating the motion at the point D , which resolve into two others FH , HD , the former FH parallel to the lever , and the latter HD perpendicular to it .
Let C be the point about which the bodies would have begun to revolve , had the force HD only acted , and which may be found by Prop. I : and suppose in this case mgn to have been the next position of the lever after the commencement of the motion , or that the bodies A , B , and center of gravity G , had been carried to m , g and n respectively .
But as the force FH acts at the point D at the same time in the direction of the rod , if we take Gq : Gg as FH : HD , then whilst the center or gravity would have moved from G to g in consequence of the force HD , it will by means of the force FH be carried in the direction of the lever from G to q , and also every other point of the lever will be carried in the same direction with the same velocity ; take therefore Ap and Br each equal to Gq , and complete the parallelograms Aa , Gw and Bb , and the bodies A , B , and center of gravity G will , at the end of that time , be found at a , b and w respectively , and awb will be the position of the lever .
Now it is evident , that C is not the point about which the bodies begin to revolve , for ( considering the lever to be produced to C ) that point must have moved over a space Cc equal to Gq , when the lever is come into the position awb : draw CO perpendicular to CB , and GO perpendicular to Gw , and O will be the center of rotation at the commencement of the motion .
For conceive CO to be a lever , then the lever ABC has a circular motion about C , whilst that point is moving from C to c , and consequently the point O is carried forward in a direction parallel to Cc by this motion ; but as the lever CO is carried by a circular motion about C in a contrary direction , it is evident that that point of the lever CO must be at rest where these two motions are equal , as they are in contrary directions .
Now the velocity of C in the direction Cc : velocity of G about C : : Gq : Gg : : ( by sim .
triang .
) CO : CG , and the velocity of the point G about C : velocity of the point O about C : : CG : CO ; hence ex aequo the velocity of C in the direction of Cc , or of O in the direction OP parallel to Cc , is equal to the velocity of the same point O in a contrary direction arising from its rotation about C , and consequently O being a point at rest , must be the center of rotation in ipso motus initio .
Also , because ma is equal and parallel to nb , ab must be equal and parallel to mn , therefore the angular velocity is just the same as if the force FH had not acted .
The center O of rotation at the beginning of the motion being thus determined , every thing relative to the motion of the bodies , after they are at liberty to move freely , may be determined as in the preceding Propositions .
Cor. 1 .
Hence it appears , that whatever be the magnitude or direction of the force communicating the motion , or the point at which it acts , the center of gravity will move in a line parallel to the direction of the force , for the triangles FHD , Gqw being similar , Gw must be parallel to ED .
Cor. 2 .
The same is manifestly true for any number of bodies ; for let ( fig. 5 .
) E be a third body , and conceive it to be connected with the other two bodies A and B in their center of gravity G ; then if FD represents the force acting at the point D , it is evident from the last Corol .
and the second Prop. that the center of gravity moves with the same velocity and in the same direction , as if the same motion had been communicated at G in a line RG parallel to FD , and that the center of gravity has the same velocity communicated to it , as if the two bodies had been placed at G ; conceive therefore the bodies A and B to be placed at G , and let the force act at D , and then from the last Corol .
the center of gravity g , of the three bodies , will move in a line parallel to the direction of the force communicated .
In the same manner it may be proved for any number of bodies .
SCHOLIU M. The method here made use of to determine the point of rotation in ipso motus initio , when a single force acts at any point D , may be applied , when any number of forces act at different points at the same time .
For let ( fig. 1 .
) ~ , ~ , ~ , &amp;c. represent the forces acting on the lever at the points D , E , F , &amp;c. respectively ; then from the same principles the effect of all the forces on A : the effect on B : : ~/ AD + ~/ AE + ~/ AF + &amp;c. : ~/ BD + ~/ BE + ~/ BF + &amp; C. which quantities put equal to P and Q respectively , and then P/ A : Q/ B : : Am : Bn : : AC : BC , from whence it appears , that ( putting GC + GA = AC and GC GB = BC ) the distance GC =AXQX AG +BXPX BG/ B XPAX Q. The same conclusion might have been deduced from this consideration ; that if any number of forces act on a lever , the effect on any point of that lever is just the same as if a force , equivalent to the sum of these forces , had acted at their common center of gravity , find therefore their common center of gravity , and conceive a force equivalent to them all to be communicated to that point , and the Problem is reduced to the case of the first Proposition .
If any of the forces had acted on the opposite side of the lever , such forces must have been considered as negative .
If there be any number of bodies placed on the lever , and a single force acts at D , it will appear from the same principles that the point C , about which they begin to revolve , will be the point of suspension to the center of percussion D ; and the same conclusion will be obtained , if the bodies be not situated in a straight line .
As a direct investigation , however , is always to be preferred to conclusions drawn from induction , it may be thought proper , before we apply any of the foregoing principles to the case of the action of bodies upon each other by impact , to show how such a direct investigation to determine the point about which a body , having a motion communicated to it , begins to revolve , may be obtained ; previous to which , however , some further considerations are necessary .
PRO P. IX .
If a force acts upon a body in any given direction not passing through the center of gravity ; to determine the plane of rotation , the direction in which the center of gravity begins to move , and its motion after .
Conceive a plane AyBZ ( fig. 6 .
) to be supported upon a line AB passing through its center of gravity G , and suppose a force to act at any point D in that line , and in a direction perpendicular to the plane ; then it is manifest , that such a force can give the plane no rotatory motion about A B. Imagine now the support to be taken away whilst the force is acting at D , then it is evident , that as the plane had no tendency to move about AB as an axis , and the taking away of the support can give it no such motion , it will , by Cor. 2 .
Prop. VIII .
begin its progressive motion in the direction in which the force acts ; and as the force is supposed not to act at the center of gravity , it must at the same time have a rotatory motion about some axis , which , as it has no motion about A B , must lie somewhere in the plane , and perpendicular to A B ; and consequently in ipso motus initio the plane of rotation must be perpendicular to the plane AyBZ .
Let LCM , perpendicular to AB , be the axis about which the plane begins to revolve , and p , q be two equal particles of the plane similarly situated in respect to AB , also qb , pa perpendicular to LCM .
Now the centrifugal force of p , or its force in the direction ap is pxap , and that of q in the direction bq is qxbq ; to determine now how these forces will affect the motion of the plane , we may observe in the first place , that the force pxap , acting at a in the plane , must tend to give it a motion about an axis perpendicular to the plane ; but as an equal force qxqb acts at q to give it a motion in a contrary direction , it is evident that the two forces will destroy each other , so far as they tend to generate any motion in the plane about an axis perpendicular to it ; and hence it is manifest , that if the parts of the plane AyB , AZB , be similar , and similarly situated in respect to AB , the plane , after the commencement of the motion , will have no tendency to revolve about an axis perpendicular to it .
Also , as the centrifugal force of each particle acts in a direction parallel to AB , it can give the plane no tendency to revolve about that line as an axis , and consequently the plane of rotation will be preserved as in ipso motus initio .
Conceiving therefore the plane on each side the line AB to be similar , and similarly situated , suppose another plane to be fixed upon this , whose parts on each side AB are similar , and similarly situated , and the force to act as before , then it is manifest , that as each plane endeavours to preserve the same plane of rotation , the two planes connected will also continue to move in the same plane of rotation , for the action of one plane on another , on each side the plane of rotation , being equal , cannot tend to disturb the motion in that plane ; and as this must be true for any number of planes thus similar and similarly situated , it is evident , that if a force should act upon a body , and each section , perpendicular to the direction of the force , should be similar on each side the plane passing through the direction of the force , and the center of gravity of the body , that that plane would be the plane of rotation in which the body would both begin and continue its motion .
It appears also from what has been proved , that if every section on each side that plane had not been similar , the plane of rotation would not necessarily have continued the same after the commencement of the motion .
Hence all bodies , formed by the revolution of any plane figure , will have the axis about which they were generated , a fixed axis of rotation ; to determine , however , every other axis of a body about which it would continue to revolve , would be foreign to the subject of this paper .
Supposing therefore the plane of rotation to continue the same ( for in this paper I mean to confine my enquiries to such cases ) imagine all the particles of the body to be referred to that plane orthographically , which supposisition not affecting the angular motion of the body , the centrifugal force of all the particles , to cause the body to revolve about an axis perpendicular to that plane , will remain unaltered .
Let LMNO ( fig. 7 .
) be that plane , and suppose a force to act at A in the direction PA lying in the same plane , which produce until it meets LN , passing through the center of gravity G , perpendicularly in D ; then by Cor. 2 .
Prop. VIII .
the center of gravity G will begin its motion in a line parallel to PA , or perpendicular to LN ; and consequently the center C , about which the body begins to revolve , must lie somewhere in the line LN .
Now the centrifugal force of any particle p is pxpc ; let fall pa perpendicular to LN , then the effect of that force at C , in a direction perpendicular to LN , will be pxpa , and in the direction CL it will be pxCa ; but as the sum of all the quantities pxpa=O , and the sum of all the quanties pxCa=the body multiplied into CG , it follows from the same reasoning as in Prop. III .
that the point G will continue to move in a direction perpendicular to LN ; and also , as the forces pxca act in a direction perpendicular to that in which the center of gravity moves , its motion must be continued nued uniform .
In the following Propositions , therefore , we suppose the axis of the body , after the commencement of the motion , to continue perpendicular to the plane passing through the direction of the force , and the center of gravity of the body , and that the body itself is orthographically projected upon that plane ; also in the case of the action of two bodies on each other , the plane passing through the direction of the striking body and point of percussion is supposed to pass through the centers of gravity of each body ; that the axis of each body after it is struck continues perpendicular to that plane , and that each body is reduced to it in the manner above described .
PRO P. X. To determine the point about which a body , when struck , begins to revolve .
Let LMNO ( fig. 7 .
) represent the body , G the center of gravity , and PA the direction of the force acting at A , which produce till it meets LN , passing through G , perpendicularly in the point D ; draw pb perpendicular to pC , on which ( produced if necessary ) let fall the perpendicular Dw ; C being supposed the point about which the body begins to revolve , and which , from the last Proposition , is somewhere in the line LN .
Because the body , in consequence sequence of the force acting at D , begins to revolve about C , and consequently if immediately after the beginning of the motion a force were applied at D equal to it , and in a contrary direction , the motion of the body would be destroyed , it is evident , that the efficacy of the body revolving about C , to turn the body about D , should any obstacle be opposed to its motion at that point , must be equal to nothing ; for were it not , the body , when stopped at D , would still have a rotatory motion about that point , and consequently two equal and opposite forces applied at D would not destroy each others effects , which would be absurd .
Now the force of a particle p , in the direction pw , being pX pC , its efficacy to turn the body about the point D is pX pC X Dw ; but by sim .
triang .
Dw : Db : : aC : pC , ~ Dw = Db X aC/ pC , and consequently the efficacy to turn the body about D=pX Db X aC =pX Ca X DC Cb =pX Ca X DC-p X PC2 ; hence the sum of all the pX Ca X DC the sum of all the pX PC2 = 0 , and consequently CD = sum of all the pX PC2/ sum of all the pX Ca , therefore D is the center of percussion , the point of suspension being at C. Cor. From this and the preceding Proposition it appears , that every thing which was proved in Prop. V. VI .
VII .
holds here also in the case of the action of one body on another .
PRO P. XI .
Let a body P ( fig. 8 .
) moving with the velocity V , strike the body Q at rest in the point A , and in a direction AD passing through the center of gravity of the striking body ; to determine the velocity of each body after the stroke , supposing them to be elastic .
The solution of this Proposition depending on the same principles as that of Prop. III .
we shall have , putting v equal the velocity of the center of gravity G after the stroke , on supposition that the bodies were nonelastic ( DGC being supposed perpendicular to AD , and C the point about which the body Q begins to revolve ) VXPX CD =vXPX CD2/ CG +vX sum of all the pX Cp2/ CG , and consequently v=VXPX CD X CG/ sum of all the pX pC2 +PX CD2 ; but it is well known , that the sum of all the pX pC2 = CG X CD X Q , and hence v=VXPX CG/ Q X CG +PX DC , and therefore if the bodies be supposed elastic , we have 2P XVX CG/ Q X CG +PX DC for the velocity of the center of gravity G after the stroke .
Now to determine the velocity of P , we have PXVX CD/ Q X CG +PX DC equal its velocity after the stroke from single impact , and consequently VPXVX CD/ Q X CG +PX DC =QXVX CG/ Q X GC +PX DC is the velocity lost by P from simple impact ; hence if the bodies be elastic , 2xQxVx CG/ Q x GC +Px DC will be the velocity lost by P if elastic , and consequently the velocity of P after the stroke =V2xQxVx CG/ Q x GC +Px DC =Px DC Qx GC/ Q x GC +Px DC x V. Cor. 1 .
If the direction AD passes through G , then CG being equal to CD , we have 2PV/ Q +P= Q'S velocity , and P Q/ P +QxV= P'S velocity , which is well known from the common principles of elastic bodies .
Cor. 2 .
If Px DC =Qx GC , Or P : Q : : GC : DC , then will the body P be at rest after the stroke .
Cor. 3 .
If Q were infinitely great , the velocity of P after the stroke would be =V as it ought , for P would then strike against an immovable obstacle .
Cor. 4 .
Whatever motion Q gains from the action of P , it would lose , if , instead of supposing P to strike Q , Q were to move in an opposite direction , and strike P at rest with the same velocity with which P struck Q ; in such case , therefore , the velocity of Q after the stroke would be V 2P x GC x V/ Q x GC +Px DC =Q 2P x GC +Px DC/ Q x GC +Px DC x V. Cor. 5 .
Hence if P be infinitely great , or Q be supposed to strike an immovable object , its velocity after the stroke will be = DC 2CG/ DC x V : hence when DC = 2GC , the body Q will have no progressive motion after the stroke , but would in such case , if P were immediately taken away , continue to revolve about a fixed axis .
It may also be observed , that when DC is greater than 2GC , or the velocity of Q is positive , that , because it is impossible for Q to continue its progressive motion , it is only to be understood , that if immediately after the impact the body P were removed , the body Q would then proceed with such a velocity .
Cor. 6 .
Suppose the bodies to be non-elastic , and let M be the magnitude of a body placed at D , which , being acted upon by P , shall have the same velocity generated as was before generated in the point D of the body Q ; then by the common rule for non-elastic bodies , the velocity of M after the stroke will be Px V/ P + M , and hence Px V/ P +M=PxVx DC/ Q x CG +Px DC , consequently M=Qx GC/ DC .
Cor. 7 .
If a given quantity of motion were communicated to any point of the body Q , the progressive motion of that body after the stroke would be the same .
For suppose the magnitude of the body P to be diminished sine limit , and its velocity to be increased in the same ratio , then , because PxVx CD/ Q x CG +Px DC ( which is the velocity of P after the stroke , if the bodies be non-elastic ) = ( because P is infinitely small ) PxVx CD/ Q x CG , the velocity of P after the stroke from simple impact is finite , consequently its motion must be infinitely small , and therefore P must have communicated all its motion to Q : now in this case the velocity of Q ( = PxVx CG/ Q x CG +Px CD ) =Px V/ Q , which quantity is independent of the place where the force acts ; in the same manner it would appear if we had supposed the bodies elastic .
PRO P. XII .
Supposing every thing given as in the last Proposition , except that the direction AD does not pass through the center of gravity g of the striking body ; to determine the velocity of each body after the stroke .
Let AD ( fig. 9 .
) be produced to meet FgO passing through g , the center of gravity of the striking body , perpendicularly in F , and suppose O to be the point of the body P which is not disturbed by the action of P on Q : now it appears from Cor. 6 .
Prop. XI .
that if both bodies were non-elastic , and a body equal to Qx CG/ CD were placed at D , the velocity of that body , from the action of P , would be equal to the velocity of the point D of the body Q ; for the same reason , therefore , it appears , that if , instead of supposing P to strike Q in the direction F A , a body equal to PX GO / FO were to strike Q at the same point and in the same direction ( which direction is supposed to pass through the center of gravity of that body ) the effect on Q would be the same ; hence , if in the quantity VXPX CD/ Q X GC +PX DC , which from the last Prop. expresses the velocity of the point D after the stroke , on supposition that the bodies are nonelastic , we substitute for P , a body equal to PX GO/ FO , we shall have VXPX DC X gO/ Q X GC X FO +PX gO X DC for the velocity of the point D from the action of P ; and consequently 2XVXPX GC X gO/ Q X GC X FO +PX gO X DC = the velocity of the center of gravity G of the body Q , after the stroke , if the bodies be perfectly elastic .
To determine now the velocity of the striking body , let Of , perpendicular to Og , be the space described by the point O in the first instant of time after the stroke , which , as that point is not disturbed by the action of the bodies on each other , may represent the velocity of P before the stroke , and let Fb represent the velocity of the point F after the stroke ; join fb , and draw gd perpendicular to Og , and then will gd represent the velocity of the center of gravity g of the striking body after the stroke .
Draw fc perpendicular to FA , and produce gd to meet fc in e ; now the velocity lost by P at the point F by simple impact being equal to VVXPX DC X gO / QX GC X FO +PX gO X DC =VXQX GC X FO / QX GC X FO +PX gO X DC , we shall have bc the velocity lost by the point F , on supposition that the bodies are perfectly elastic ( supposing of to represent the value of V ) equal to 2XVXQX GC X FO / QX GC X FO +PX gO X DC , and therefore by sim .
triang .
fc ( FO ) : cb : : fe ( Og ) : ed = 2XVXQX GC X gO / QX GC X FO +PX gO X DC = the velocity lost by the center of gravity g , and hence V2XVXQX GC X gO / QX GC X FO +PX gO X DC = VXQX GC X FO +VXPX gO X DC 2XVXQX GC X gO / QX GC X FO +PX gO X DC = the velocity of P after the stroke .
Now , as it appears from Prop. IX .
that the progressive motion of a body , when left to move freely , continues uniform and in the same direction , it follows , that the expressions for the velocities of each body in the first instant after the stroke , both in this and the preceding Proposition , will represent the uniform progressive velocities with which the bodies will continue to move , and consequently the place of each body , at the end of any given time after impact , may easily be determined .
Cor. 1 .
If the direction FA passes through g , then FO and gO becoming infinite , we shall have 2XVXPX GC / QX GC +PX DC for the velocity of Q , and VXPX DC VXQX GC / QX GC +PX DC for the velocity of P , agreeable to what was proved in the last Proposition .
Cor. 2 .
Hence the point about which P begins its rotatory motion may easily be found ; for produce ( if necessary ) fb and OF to meet in a , and a will be the point required ; and by sim .
triang .
bc ( = 2XVXQX GC X FO / QX GC X FO +PX gO X DC ) : cf : : fO ( = V ) : Oa =QX GC X FO +PX gO X DC / 2XQX GC , and hence Fa =PX gO X DC QX GC X OF / 2XQX GC .
Cor. 3 .
If , instead of supposing Q to have been at rest , it had been moving forward in a direction parallel to that of the body P , with the velocity v , the motion of each body after the stroke may easily be determined : for considering P as acting upon Q with the velocity V v , we shall have by this Proposition ( putting 2M = 2P X GC X GO / QX GC X FO +PX gO X DC ) VvX 2M = the velocity communicated to G , therefore v+VvX 2M = the velocity of Q after the stroke : also VvXMX CD / CG = the velocity gained by the point D from simple impact , and consequently the velocity of that point after =v+VvXMX CD/ CG , hence VvVvXMX CD CG = the velocity lost by P at the point F from simple impact , therefore P'S velocity after the stroke =VVvVvXMX CD / CG X 2gO / FO .
In the same B ADGEFBCAG n gw m gK FH O P Fig. 4 .
Fig. 5 .
FH ABCFR pGqrc D ~ o gADG w m a Fig. 7 .
M Q~ w LDaGCN b DGC p AO Ap P Fig. 3 .
B TIE A D WGX B RC Fig. 6 .
YL Fig.5 .
E qb DGC AB FRg pa DGBZM Q Fig. 8 .
DGC CA fec db OgFa Fig. 9 .
P manner it might have been determined , had Q moved in an opposite direction .
Cor. 4 .
Hence also we may easily determine the motion of each body after the stroke , supposing Q had not been moving in a direction parallel to the motion of P , by resolving Q 's motion into two parts , one parallel to the motion of P , and the other perpendicular ; and finding by the preceding what would be the effect of the parallel motions , and then compounding Q 's motion , after the stroke from that consideration , with the motion it had in a direction perpendicular thereto before the stroke .
Cor. 5 .
The point a of the body P will describe ( when that body after the stroke has any progressive motion ) the common cycloid .
Cor. 6 .
Hence , therefore , the times of the revolutions of each body may be determined as in Prop. VI .
Cor. 7 .
If the bodies had any rotatory motion before impact , every thing relative to the motion of the bodies after the stroke might have been determined from the

