Memoir on the Automorphie Linear Transformation of a Bipartite Quadric Function .
By ARTHUR CAYLEY , Esq. , F.R.S. Received December 10 , 1857 , -Read January 14 , 1858 .
THE question of the automorphic linear transformation of the function a9+y2+z2 , that is the transformation by linear substitutions , of this function into a function e2+y2+z2 of the same form , is in effect solved by some formulae of EULER'S for the transformation of coordinates , and it was by these formulae that I was led to the solution in the case of the sum of n squares , given in my paper " 1 Sir quelques proprietes des determinants gauches* .
" A solution grounded upon an a't-priori investigation and for the case of any quadric function of n variables , was first obtained by M. HERMIT in the memoir " Remarques sir un Memoir de M. CAYLEY relatif aux determinants gauchest .
" This solution is in my Memoir " Sir la transformation d'une function quadratique en ellememe par des substitutions lineaires$ , " presented under a somewhat different form involving the notation of matrices .
I have since found that there is a like transformation of a bipartite quadric function , that is a lineo-linear function of two distinct sets , each of the same number of variables , and the development of the transformation is the subject of the present memoir .
1 .
For convenience , the number of variables is in the analytical formulae taken to be 3 , but it will be at once obvious that the formulaz apply to any number of variables whatever .
Consider the bipartite quadric ( a , b , c fie , ZI ?
X , YJ Z ) |a ' , 6 ' , c'| a , l ' , b " which stands for ( ax +by +cz)x +(a'x + Vi + cz)y + ( a~x+ bay + 'z)z , and in which ( x , y , z ) are said to be the nearer variables and ( x , y , z ) the further variables of the bipartite .
It is clear that we have ( a , b , c )Xx , y , z}x , y , Z ) a , a ' , a " fix , y , z , '~x , y , z ) |atI b ' , el |b ~b ' bit| ari ' , b " , c " -c , c " and the new form on the right-hand side of the equation may also be written ( tr.( a , b , c )x , y , zIx , y , z ) , al bi C~ a " , 6 " , c " ' that is , the two sets of variables may be interchanged provided that the matrix is transposed .
3 .
Each set of variables may be linearly transformed : suppose that the substitutions are ( * , Y , z)-( I , In n IX , , , Y9Z , ) 1 ' , in ' , n ' and ( x , y , z)=( 1 , 1 ' , 1 " XS y , Z , ) m , m ' , Im " n , niI nit Then first substituting for ( x , y , z ) their values in terms of ( x , , Y , Z ) , the bipartite becomes ( ( aI bI c 3C 1,5 mg , n1 )x , y , xXx , yl Z ) a ' , b ' , c ' 1 ' , in ' , n ' Represent for a moment this expression by ( A , B , C , X3x , , y , , z , Xjx , y , z ) , A ' , B ' , C ' , A " , B " , C " , then substituting for ( x , y , z ) their values in terms of ( x , , y , , Z , ) , it is easy to see that the expression becomes d re m , n XtAi Bi C or tri , we zoinxed , a l , a it1 ' m ' , 111 A ' , BlI C ' 1"l , ml/ nt"l , noB " ' C"l of the substitutions , |x , a " , , b " , c " | | , m " , n a " , b &amp; || 1 " m n , n " t that is , the matrix of the transformed bipartite is obtained by compounding in order first or furthest the transposed matrix of substitution of the further variables , next the matrix of the bipartite , and last or nearest the matrix of substitution of the nearer variables .
4 .
Suppose now that it is required to find the automorphic linear transformation of the bipartite ( a , b , cx ]( , Y , z:x , y , z ) , a 6 " , c " or as it will henceforward for shortness be written , ( QX , y , z)1x , y , z ) .
This may be effected by a method precisely similar to that employed by M. HERMIT for an ordinary quadric .
For this purpose write as+x , ,=2 , y z+z1=2K , x+x , =23- , y+y=2H1 , z+z1=2Z , or as these equations may be represented , ( + 1 , , y+y , , Z+zI)2( ) we ought to have ( Q](26-x , 27y , 24-xzXC2:-_X5 2Hy , 2Zz)-(Q](x , y , z](x , y , z ) .
5- .
The left-hand side is 4(Q:(t5 , , L= , HvZ)-2(Q~ty , yz3Ct , 5H.Z)-2(QE~5 , tt x , yz)+(Q39 y , z3Ex , y , z ) , and the equation becomes 2(01t4 , 2 , 43C.- : H. Z ) --(Qiy tI )(~ t 3xX yXz)=OX or as it may be written , ( Q6 q 43tH .
Z-Qy t , H. Z ) +(01~5,4 4 3C:E H5 Z)(01~5'2 g 43tg y z5 Z)J or again , ( Q _-* 4-z , ( :E3t , H , Z)| and it is easy to see that the equation will be satisfied by writing ( Q)C4 _x , -Y5 U-z)=(Y 18 , , 5 ( trQ38 H-xI-ys Z-z)=-(tr .
YK= , H5 Z)5 where Y is any arbitrary matrix .
In fact we have then ( Ql6 Ax5 8-y , Kzj0 , Elf Z)I(36 , 5 H5g , I Z)9 ( tr .
Qzx , Ho-y , Z z38 , K ) ( tr .
YjZ , H. Z3j , , c-( Y~ji 8 1:4X H. Z ) , and the sum of the two terms consequently vanishes .
6 .
The equation gives ( Q Yip , 4 ) 4)-M(Q38 # Y4 % ) and we then have ( Q+Y6 8 , ) ( QS2Xr , y , , Z , ) In fact the two equations give ( 2Q , 4 , )-(Q:(+$ , , Y+Yo:+X)~ or what is the same thing , 2(6 ) n ) 4)=(x+$x ) , , Y+ , + : , )t which is the equation assumed as the definition of ( 6 , a , t ) ; and conversely , this equation , combined with either of the two equations , gives the other of them .
7 .
We have consequently M6 4 ) t)((Q+Y)1Q3J( , Y , ZI , ) and thence 8 .
But in like manner the equation ( tr .
S JXt : -x , HI y , Z -z)= ( tr .
yj : : H. Z ) gives ( tr .
Q+Yj= , H. Z)=(tr .
Qicx , y , z ) , and we then obtain ( tr .
Q-..Yip , ~ H. Z)=(tr .
Q3Cx , , ye z , ) 9 .
In fact these equations give ( tr .
2QXZ- , H , Z)=(tr .
Qlx+x , , y+y1 , z+z , ) , or 2(= , H , Z)==(x+x , , y+y , , Z +z)i and conversely , this equation , combined with either of the two equations , gives the other of them .
We have then and thence ( X ) ye z)=((tr Qltr .
Q+Y)(tr .
Q-Y)-Itr .
, Q ' ( x , yes z , ) .
10 .
Hence , recapitulating , we have the following theorem for the automorphic linear transformation of the bipartite ( Q(Xt , y7 z)x , y5 Z)4 viz. Y being an arbitrary matrix , if ( , q z)=(Q-1(Q-Y)(Q+Y)'QXX , , y1 , z ) , ( x9 y , )=((tr .
Q)-'(tr .
Q+)(tr .
Q-Y)'ltr .
Q:Cx , , ye z , ) then ( Q(X , Y. z3Xx , y , z ) =(Q3X y , , zXx , y , z ) which is the theorem in question .
11 .
I have thought it worth while to preserve the foregoing investigation but the most simple demonstration is the verification 'a osterzori by the actual substitution of the transformed values of ( x , y , z ) , ( x , y , z ) .
To effect this , recollecting that in general tr .
( A'-)=(tr .
A)- ' and tr .
ABCD=ztr .
ID .
tr .
C. tr .
B. tr .
A , the transposed matrix of substitution for the further variables is and compounding this with the matrix Q of the bipartite and the matrix of substitution for the nearer variables , the theorem will be verified if the result is equal to the matrix Q of the bipartite , that is , we ought to have or what is the same thing , 2(Q-YY'(Q+ -Y)Q'l(Q_-Y)(Q+Y)-YQ=Q ; this is successively reducible to Q'(Q+Y)Q'1(Q0-Y ) Q-Q(Q -Y)Q-(Q+Y ) ( 1+n-ly)l_ Q_1Y)(1 _ Qy)(I_ Q 0Y which is a mere identity , and the theorem is thus shown to be true .
12 .
It is to be observed that in the general theorem the transformations or matrices of substitution for the two sets of variables respectively are not identical , but it may be ' required that this shall be so .
Consider first the case where the matrix 0 is symmetrical , the necessary condition is that the matrix Y shall be skew symmetrical ; in fact we have then tr .
Q=Q , tr .
Y=-Y , and the transformations become ( XY , Z ) ( ( Q 0Y ) 2 its y , Z).~~\ 2 , \/ - , A - , I , _ _ MiR .
A. CAYLEY ON THE AUTOMORPHIC LINEAR which are identical .
We may in this case suppose that the two sets of variables become equal , and we have then the theorem for the automorphic linear transformation of the ordinary quadric viz. Y being a skew symmetrical matrix , if ( X7 y , z)=(Q '(Q-Y)(Q+Y ) 'Qlx ) Y Z ) , then ( QXa , y , z)2=(QX , yz )2 .
13 .
Next , if the matrix ? ? ? ?
be skew symmetrical , the condition is that the matrix Y shall be symmetrical ; we have in this case tr .
Q =Q , tr .
Y== Y , and the four factors in the matrix of substitution for ( x : , y , z ) are respectively Q -( -Y ) , -(Q+Y ) and -0 , and such matrix of substitution becomes therefore , as before , identical with -that for ( x , y , z ) ; we have therefore the following theorem for the automorphic linear transformation of a skew symmetrical bipartite ( QjX$ , Y , AJx , Y9 a ) , when the transformations for the two sets of variables are identical , viz. Y being any symmetrical matrix , if ( x , y , z)=(Q'(Q-Y)(Q+Y)-'Qjx , y1 , z , ) , then ( Q:(X , Y. Z3(X9 Ye z)=(Q:(X19 yl , %3X , Ye Z1 ) .
14 .
Lastly ; in the general case where the matrix Q is anything whatever , the condition is --'y= ( tr .
Q)- ' tr .
Y for assuming this equation , then first Q0'(Q-Y)-(tr .
Q)-'(tr .
Q+Y ) , and in like manner But we have 1=(tr .
0)'1(tr .
Q-Y)(tr .
Q-Y ) tr .
0 , and therefore , secondly , ( 0+Y)-'Q-(tr .
0-.-Y)-tr .
0 ; and thence Q-1( -Y)(f2+Y)-1 =(tr .
02)'(tr .
0+Y)(tr .
Q-Y)tr .
0 , or the two transformations are identical .
15 .
To further develope this result , let Q- ' be expressed as the sum of a symmetrical matrix Q0 and a skew symmetrical matrix Q. , and let Y be expressed in like manner as the sum of a symmetrical matrix Q0 and a skew symmetrical matrix Ye We have then Q_1 ~~=QO+Q , ( tr .
Q)-= tr .
( Q- ' ) =Q0-Q , , Y -Yo+y tr .
Y =Yo- , and the condition , 0 1Y -(tr .
Q ) -'tr .
Y , becomes NQO+ Q)(Y + YI ) A ( Q Q , ) ( YO Y. ) that is , QoY0+QY , = 0 , and we have or as we may write it , Yo= -(-{Q-1+tr .
Q`J } ) l ~tr .
Q-1j )Y and thence Y= 2 -lt ')-l Qatr .
Q-11)Yl+Y7 where Y is an arbitrary skew symmetrical matrix .
16 .
This includes the before-mentioned special cases ; first , if 0 is symmetrical , then we have simply Y=Y , , an arbitrary skew symmetrical matrix , which is right .
Next , if O is skew symmetrical , then Y= -0-Q-0Y , +Y , , which can only be finite for Y 0 , that is , we have Y= -0I'Q-I0 , and ( the first part of Y being always symmetrical ) this represents an arbitrary symmetrical matrix .
The mode in which this happens will , be best seen by an example .
Suppose Q-1 -XA H+v ) , tr .
Q`=( A , 1-v | H-P aB |jH +v , B and write X , -0 0 then we have T=-( AH )( v)( 0 , 0 )+( 0 , )| H. B j- , O0 f |-0 , 0 L- , 0o| AB-H2 ( -B .
H +( 0 , d ) H1 -A j-oioj Y sBO -YITO )( 0 , 0 ) AB-H2AB-I 2+8 [ oI -____ vAG AB-R !
H2 AB-H2 When 0 is skew symmetrical , A , B , H vanish ; but since their ratios remain arbitrary , we may write zA , zB , xH for A , B , H , and assume ultimately X=0 .
Writing xd in the place of 0 , and then putting X=O .
the matrix becomes vB0 -vH ) AB-HI2 AB-H2 vH_ vAG AB-H2 AB--H2 which , inasmuch as A : , B : 0 , and C : 0 remain arbitrary , represents , as it should do , an arbitrary symmetrical matrix , 17 .
Hence , finally , we have the following Theorem for the automorphic linearNtrans .
formation of the bipartite quadric , ( 0-J-T ) Y , zXx , y , 4 when the two transformations , are identical , viz. if Y , be a skew symmetrical matrix , and if then 'if 811G ~~ ( X 1IY1 ) W~e have ( 0_ _ ( i Y21 OXrI Q3 -Z})( { Q:(_ rQo } Z)l,5 and in particular , if ?
" is a symmetrical matrix , then Y is an arbitrary skew symmetrical matrix ; If 0 ?
i-sa skew symmetrical matrix , then Y 'is an arbitrary symmetrical matrix .

