[ 3 ' 1 II .
On the Solution of Linear Differential Equations .
By CHARLES JAMES HARGREAVE , Esq. , B.L. , F.R.S. , Professor of Jurisprudence in University College , London .
Received June 10 , -Read June 17 , 1847 .
I. General Theorems of the Calculus of Operations .
IF the operation of differentiation with regard to the independent variable x be denoted by the symbol D , and if p(D ) represent any function of D composed of integral powers positive or negative , or both positive and negative , it may easily be shown , that p(D{-x~u=-A , ( D)u+A'xwp'(D)u+ HxZ.p."(D)u+,3 111.P11(D~ , ) and that t.,4(D)u== ; ( D ) { px.u } -+'(D ) { 'x.u } +24"(D ) { p`x.u } 4+"'(D){p"'x.u}+.* ( 2 .
) and these general theorems are expressions of the laws under which the operations of differentiation , direct and inverse , combine with those operations which are denoted by factors , functions of the independent variable .
It will be perceived that the right-hand side of each of these equations is a linear differential expression ; and whenever an expression assumes or can be made to assume either of these forms , its solution is determined ; for the equations p(D){Zx.u}=P and px.4(D)u=P are respectively equivalent to u=(4x)-'{p(D)}"-P and u={4(D)}-'((px)-'P ) .
The formulae ( 1 .
) and ( 2 .
) indicate true propositions whenever they are interpretable ; that is , whenever p(D ) and +(D ) are capable of being expressed in integer powers of D. In conformity with recognized principles of reasoning , when the subjects of the process are regarded merely as symbols , we may assume that these propositions are true generally ; and we shall therefore not hesitate to pronounce any interpretable result derived from the free use of these theorems true , although the intermediate steps of the process are not capable of a rational interpretation .
Bearing these remarks in mind , it will be seen , by an inspection of the above equations , that if in ( 1 .
) D be written for x , and -x ( or t ) be written for D , we obtain pt.4i(D')uu =(DI ) { pt.u } -.4(D ' ) { p't.u } + 2.k"(D ' ) { 1"t.u .
do D'n dentin the operation.''1 _ MThis eqainisietcl nfr wt 2 ) and is therefore true ; and the correctness of the result thus derived from this interchange of symbols leads to the inference , that if in any linear differential equation capable of ' being reduced to the form ( 1 .
) or ( 2 .
) , and its symbolical solution , x be changed into D and D into x , we shall , obtain another form accompanied by its symbolical solution .
Possibly the form so obtained and its solution may not be interpretable ; but in every case in which they are interpretable , they will be found to be true ; and if , by any transformation , meanings can be attached to those forms which appear to be unintelligible , they also will be found to be true .
It is an essential condition to be observed in all operations in which this process is used that the solutions are to be preserved in a symbolical form ; or , in other words , that the operations are not to be performed or suppressed .
It would manifestly be a source of error to write zero for ( Zx').0 , if in a subsequent stage x is to be converted into D. The process may be conveniently exemplified by applying it to the general equation of the first order , Px .
Du+ , x.u=X ; of which the solution , ( the processes being preserved , ) is { ( _(pD ) 1X ) } , or u-xD If we make the proposed conversions , we have for the solution of -p(D){xu}+4'(D)u=X.(3 .
) But equation ( 3 .
) , by ( L. ) , is equivalent to xq)(D)u+((p'(D)--+(D))u=-X1or X0 .
Let P'(D)-4P(D)=k(D ) , or +(D)-@'(D)-XD .
Then X(D ) ) dD=log p(D ) -f , ( dD ; and the solution assumes the form = ( 4 &gt; ( D))-fA(D)d{x ADXo } ... *(4 .
) the equation to be solved being xP(D)u+X(D)u=X0 .
This solution was first given by Mr. BOOLE in the Philosophical Magazine for February 1847 .
It indicates in a striking manner the interchange of symbols which is here proposed as a general theory ; and leads naturally to the inquiry , whether such a conversion may not be extended to other forms .
I am not prepared to assert that the considerations stated above actually establish its validity as a theoretical process ; but it possesses considerable practical utility , when applied to a subject in which the value of the result , if true , is in a great measure independent of the validity of the process .
In this paper I propose to apply the formula ( 1 .
) , by the aid of Mr. BooLE'S solution above given , to the discovery of soluble forms of linear equations with variable coefficients ; I shall also show that by the use of the conversion of symbols , many forms of solution apparently incapable of interpretation may be made to give useful results ; and I shall point out a remarkable connection between the solutions thus obtained and the solutions of the same equations in the form of definite integrals .
Amongst the cases of ( 4 .
) which are obviously and immediately interpretable , may be mentioned xp(D)u+mp'(D)u=X u =f @(D ) 3m-14x-1 { ( D)J}-mX J ; ( D ) ( D)u~m( )u=X D.(6 .
) U={ p(D ) } -Dm{x-D-mX } , ; but , as will be afterwards shown , most cases are interpretable when p(D ) and X(D ) assume the ordinary form , and consist only of integral powers .
II .
Application of these Theorems to the Solution of Equations .
I proceed then to apply equations ( 5 .
) , in conjunction with the original theorem ( I. ) , to the solution in finite terms of formis of linear differential equations .
Commencing with equations of the second degree , we have , by ( 1 .
) , { D2+ bD+c2 } ( X.u } =4'x.D2u+ ( b4nx+ 2 , 'x)Du+ ( c24'x+ b'4x+.41'x)u { 2D+b } ( %x.u } = 2.4x.Du ?
+ ( b'4xx+2#'x)u .
Consequently equations included under the form x4~x.D2U+ ( ( bx+2m ) 4ix+2x41x)Du+ ( ( c2X+ bm)4'x+ ( bx+2m ) &gt; 4'x+ x-"x)uX , o IrJ2u+ ( b+ Du+ ( C2 +)+ )u= ( 4x)'X=P , are readily soluble , the solution being u= ( ')x-'(D2+ bD +c2)mf-'{xN-(D2+ bD+c2)-mX } = ( 4x)-'(D2+bD +c2)m-l{x-'(D2+bD+ C2 ) -m(Z4 P ) } .
When X or P is zero , the solution may be reduced to the simpler form , u= ( 4x)-1(D2+bD+ c2)m-l{x-(D2+bD+c)-2.O } , which will be found to be u= ( 4~x)-1x-m{k ?
M(l + ( in-1 ) ( ,3 n &gt; + ( m.-1 ) 2~m(m+1 ) m-2~~~ m-23 m~~lmm2)\l m U6x-xnk'1+(m-1 ) ( 3 -+ *)+ksZl(n1)(}-~ -2 m-3m(m + 1 ) ( m -2 ) mM )(+2 +(m-l)m2 ( m+3(l )I2 m+ 3 ( ma)3m a and f3 being the roots of t2+bt+c2=O .
NnCCCXLvIII .
p tOF The solution here given is finite , in those cases only in which m is an integer positive or negative .
When ni is fractional , the undeveloped expression involves fractional operations .
If however b= -2c so that the roots are equal , the solution assumes the form U= ( pX )-ISex(x-2m+l+kI ) without any restriction upon the values of m. The form ( 7 .
) deserves particular attention , as it will be found to include the most remarkable of the equations of the second order , which have heretofore been integrated by artificial methods .
Thus , if 4x=s ' we have the solution of D2U+ ( b+2a+r)Du+ ( c2+ab+a2 ?
( b+2a)Z)u=P , of which the well-known equation D2u+ Du+c2u=P.(8 .
) is a particular case ; the solution being u= ( D2+c2)m-l { X-1 ( D2 ?
C2)-m(XP ) } If P=0 , this is reduced , taking the negative sign only , to u= ( D2C2)m-l { 1-(kzcx+k'e-cx ) } , which will be found to be U +{krcx ( 1-(m-1)2x+(m1)m2 mm+1)_ )+ kt ?
-^ZC 1+(nz-1 ) 2cx+ ( M -I ) 2 Xr(2 Zx)2i +f ) } and the solution is expressible in finite terms when m is a positive or negative integer. .
This equation is merely the simplest form of ( 7 .
) , and is soluble by ( 5 .
) without the aid of ( 1 .
) ; for in it 4Ax is taken to be unity .
Now let A=xn ; then the solution of D2 + 2(m ?
n ) +2 + b(mjn ) n(n-l)+2mn ) Dt+ b+--X Du+tc+ X+ -2 2-JU= P is u=x-n(D2+bD+ C2)m-l{x-'(D2+bD+c2)-n(xn+ 'P ) , } If in this b=0 and n= -2rn+I , we have another form of solution of the equation D2u .
2(m )Du+C2U=P ; namely , u =x2m-1 ( D2+ C2)m-l { x-'(D2+ c2 ) -m(x-2m+2p ) } .
If n= -m , we have for the solution of D2U +bDu+ C2m(M ' ) ) u=P u =xr(D2 +61D+ c2)m-1 { X-1 ( D2+ bD c2)-m(xmilP } ; of which the well-known equation .
D2u(m(rn2 I ) ?
2)O is a case ; the solution being u=a'm(D2+c2)m1{x-1(k sin cx+k ' cos cx ) } U=xm(D2C2)m-l { x-1 ( kecx+ktgc ) } The latter form is simply the series before found in the solution of ( 8 .
) without the factor xm .
In fact the solution of ( S. ) is the type of the solution of ( 7 .
) , when there is no second term P ; for if u=ul is a solution of ( 8 .
) , u=(4'x)-'u is the solution of D2u+ ( 2+ !
)Du + c2 + 7)uO .
The suppression of the terms containing b does not materially impair the generality of the form ; for it is well known , and follows immediately from ( 1 .
) , that p ( D ?
+ ) u=_ 2p(D ) { s2xu}j .
I have found the most convenient form of ( 7 .
) to be D2U+2QDu+ Q22+Q2+Q '_(3)UP }.(9 .
) ft= ?
-Ql ( D2 + C2)m-l { 1 ( D2 + C2 ) -m(X-(m-l)QIP ) } which are obtained by making Q= , + , , Q1 beingjfQdx .
Useful applications may be made by eliminating the second term from ( 9 .
) by a change of the independent variable from .
x to t ; t and x being connected by the equation d=--2Q , One of these applications leads to an investigation calculated to throw some light upon the limited character of the solution of RICcATI 's equation .
If in ( 9 .
) Q be taken we have as a soluble form , D ?
2+2n rDu+C2 n(n-1)-m(m-1)\ p and the elimination of the second term gives , ( making d= nd z=-( t ) dU 5Z+(2n1)-2 ( c2z 2n-1+ ( n(n1)-m(m1))z -2 ) u=R or pz2n-1 , This form , therefore , and the cognate form dv +v2+(2-n2 C2z-2-i+(n ( n1 ) ( m-l))Z-2 ) =0 are soluble , without the restriction that n must be a whole number ; but when this equation is made to assume RiCcATI'S form by equating n to m , the restriction on the values to be given to n takes effect .
If for n be written -n , we obtain the other form of the equation , viz. dv J Zn M-)=0 do+V2 + ( 2n +1 ~(Czs+l + ( n(n +1 )-(m13z)=O which is subject to the same restriction , when assimilated to RICCATI'S form .
The solution of dgu a\2 _4n is2 g+(2c1 , )2Z2ZnU1O is { x ( D2+c2)n'{x'ksin cx+k ' cos cx ) } , x being z2n-l ; and that of d -2 + ( 2 c+ I ~8+l is x2n+1(D2+c2)1{xI(ksin cx+k'cosct4 from which , general expressions for the solution of the two corresponding forms of RICCATI'S equation may be deduced , subject to proper precautions with reference to the arbitrary constants .
If we now , in a similar manner , apply the equations ( 6 .
) in conjunction with the original theorem ( 1 .
) , we shall find , making p(D)=D2+bD , that equations of the form D2u + ( b+ 2 ?
)Du +( +(b ?
J $+ r )uP are soluble ; the solution being U= ( 4x ) -(D2+ bD)'Dm { -D-m ( Z4'x.P ) } which assumes , m being integer , the form of the finite series , u= ( +~x ) -l(D + b)-{ XD_1(ax .
P ) _ ID-D2(4wZ .
P ) +2Z D ' ( 4xtP ) -2.3(m-l)(m-2)(m-3)D 4(x+x.P)+ * And again , making b=O and Q=a+-x+2mr , we have foithe integral of D2u+2QDu+ ( Q22QJ( , ; )u= P , u=x22Q1Dm I-{D m(x 2 ?
'Q1P ) } .
By processes in all respects similar , integrable forms of equations of the third and higher orders may be obtained .
For equations of the third oider , it will be found that the expressions ( 5 .
) give , suppressing b , D3u + 3(MkJ+4D2U+ ( c+ 6m $+ 31 r+x +L(J )f $+ )P xx ~~x +X ?
,1X kJfkajFx xjx4x 4x u= ( ix ) -'(D3+ cD+f)m-l{x-1(D3+cD+f)-m(x4#x.P ) } ; or D3u+3QD2u+ ( c+3(Q2+Q ' ) _3m(M2 '1)Du+ ( +cQ+Q3+Q"'+3QQ ' -3m(m-1 ) Q+(m l)n(m+1 ) )p -3 2 Q+--3-y u=azrn-Qi(D3+ cD +f)m1 { l -(D3+ cD+f ) -m ( V-m+lgQp ) } And the expressions ( 6 .
) give D3u+ ( b +j +3L )D2U+ ( c++2 ( b+ !
j !
x+3 f ?
)Du + ( C + C+ +A u= ( +xv ) ' ( D2+bD + c ) -1Dm-1Q-D-i(xix .
P ) } It is obvious , however , that the generality of the soluble forms becomes less , as the order of the equation rises .
The solutions derived from ( 5 .
) and ( 6 .
) as particular forms of ( 4 .
) , have been given in the first instance on account of their peculiar simplicity ; but more general forms are derived by the use of ( 4 .
) .
The expressions ( 4 .
) represent the solution of linear equations of any order , in the factors of which no power of x higher than the first appears .
The general form is ( at ++b)Dnu+(an-Ix+be i)Do-lzt+.**+(alx+b1)Du+(aox+bo)u=X,. .
( 10 .
) in which A~t_ bath+b._ t1+. .
+ bt +bo pt= aJt+ a_ lt ' +1. .
+ alt + ao / bo b , +A+B_ dt= t+ log ( ( tz)A(tP)B( tie )C .
) .
d..1 where a , j3 rX &amp;c. are the roots of ItO - ; and A , B , C , &amp;c. are found from the rational fraction Consequently the solution of this equation is a ( anDn + an_ , Dn'I .
+ ajD + ao ) ls ( D _)(D _ 3..(I C ?
an ( D-z)AD_0The factor &amp; aD denotes that x is to be changed into x a. and the factor SaaD denotes that x is to be changed into x+a ; but these factors may be dispensed with by making b_=0a which does not diminish the generality of the form .
If , for example , we apply this theorem to the solution of D2u+(al +bi)Du+(aoxv+bo)u=X ; we have the rational fraction bit +bo A ?
bla+bo B -1p+bo t2+a t+ao ' a3 u=(D2+aD +a)-1 ( D-Ce)A(Dp-)B(x{ ( Da ) -A(D -3)-X The performance of the operations requires that A , B , &amp;c. shall be whole numbers ; and these are the conditions under which the equations are soluble in finite terms .
If we combine with the above general form the formula ( 1 .
) , we obtain the solution of the equation ( an x+ Jbn ) Dnu+ an-1x ?
bl ?
n(ax+ be ) , + { an-2x+bn-2 +(n ] ) ( a. lx +bn 1 ) - , X+qMscn ( an + bn)V Dga }n-2U ~~Px~ ~~xF 'li~ ~ ~~~~~X )n-2 4 , x n-1ln-2 1X +an-3-V+bn-3+(n -2 ) ( an-2X+ b:+ ( n- ) 2 axnf-3gzn2 3(axb)\ ~~~~~~~~~~~~ )2~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-)(a , -1x+b 1)-x n2 ax3 + ... -X in the form U=(%x)'1(anD +an1Dt-1+.)a(D ' -(D -3)B a , ( D-e ) ( D.- ... 3 ) .
( 4@xX ) The important limitation , that A , B , &amp;c. must be whole numbers in order that the operations may be practicable , must not be overlooked ; and with reference to this point , the attention of the reader is called to the solutions by means of definite integrals given in a subsequent part of this paper .
If two or more of the roots oa , A , &amp;c. are equal , we obtain amongst the operations expressions of the form a ( D ) " ) ' which do not appear to be interpretable in finite terms ; but the corresponding solution in the form of a definite integral will apply .
III .
Solution of Equations by interchange of Symbols alone .
It has been already ol)served , that the operation or set of operations denoted by any function of D is not of itself intelligible , unless the function is capable of expansion in integer powers of D , so that fractional operations may not be introduced ; but if , by means of the transformation above indicated , the function of D becomes in result changed into a function of x , such a result is intelligible , and may be relied on as true , although the expressions introduced during the process may be purely symbolical and incapable of interpretation .
Thus U= ( D2C2)mi-{arl(D2-c2)-mX } , regarded as the solution of a(D2c2)u+2mDu= Xx is not interpretable in finite terms when m is fiactional .
Yet u= c2)m-1D{j,2_ c2)-mX } , regarded as the solution of D{(X2 c2)u}-2mxu=X , or ( X2 _c2)Du-2(m-l)xu=X , ( these forms being derived from the others by changing D into x and x into D ) , is interpretable for all values of m , and is correct .
Again , it has already appeared that the solution of D +2n a(2n(n-1)-M0(M~-1 ) ) D2U+-Du+ ( c2+ ( U 1 ) is u=xvm , -(D2+C2)m-lx l(D2+C2)-m ( n-m+lp ) } ; which is not interpretable in finite terms when m is fractional .
But if this equation be multiplied by x2 and transformed as before , we get ( X2+ c2)D2u-2(n-2)xDu+ ( ( n-l)(n-2)-m(m l))u=D2P=R suppose ; and we infer the solution to be u=Dm- , f{(X2+C2)m1-D-'f ( X2+C2)-mD-mlR } } ; which is interpretable though m be fractional , provided mn be an integer ; or the equation ( X2+c2)D2u-2axDu+b(2ab+ l)u=R is soluble when b is a whole number ; the solution being u=D-(b+1){ ( X2+c2)a-bD ( X2+c2 ) ( a4 +I)DbR } } of which a remarkable case is , ( a=O ) D2Utb2 2b ?
cU( U= e+c2=X ( suppose ) U=D(b+1'){( , V2+C2)-bD-l{(X2+C2)b-lDb((X2+c2)X ) } } .
In applying these forms great caution must be used with reference to the introduction of constants .
The processes indicated in the value of u show b+2 constants .
; which renders it necessary to determine b of them in terms of the other two by reference to the original equation .
Thus , suppose we require the integrals of the equations ( 1 +x2)D2u-2u=x , and ( l +V2)D2u -2u=a .
The form of both solutions is the same , viz. U=kD -3+1)2 whence we have D2u=k 1 2+I2 tan-x)r+kn k Du=-x tan-lx+k'x+k " uj ( ( L+X2 ) tan-axk ) +2 + ?
kx+kl " .
Verifying the first equation by these values , we find k_2 and k'-2 , the solution , therefore , is k P( +2 ) uz=((I+.V2 ) tan-'+x)-2+2(1+x2 ) .
k k'-a Verifying the second equation , we find k"= and k"'=-and the solution is k k ' a u=-((1 +x2 ) tan-'x+x ) +2 ( 1 +,2 ) If the general form of the above differential equation be divided by X2+c2 , and the second term be eliminated , by changing the independent variable from x to t , by means of the assumed equation d= ( x2+2 ) , we obtain d2u dt2 + b(2ab+ 1)R ( +c2-(2a+)uR(x2+c2)(2a+13= R0 suppose .
In order that the factor of u may be expressed in terms of the new variable , we must dt solve -= ( -2+ C2)a , and then find x2+ c2 in terms of t. For most values of a , the equation between x and t is transcendental ; but particular soluble cases may be found .
Thus , if a= then x=c tan ct , a2+c2=2ct ' and the equation becomes d2u2 d2b(b +l 1)o2 CtUR0 ; whose solution , therefore , is U=D -(bl { ( x2+ c2)(6+l)D-'{ ( X2 + c2 ) bD ( ( ( X2 + c2 ) -'Ro ) } which becomes , when Ro=-0 =k ( ^+l ) ( Z2 x being c tan ct .
The original equation is not altered by writing b for b+ 1 ; so that a particular solution may be readily deduced from the simple form u kD =(+ c2 ) .
This equation will reappear under another form in the sequel .
c2 Again , if a=2 , then c2+x2-and the equation becomes d 2_(b+ 9 ) c4U -R1 ; whose solution , therefore , is u=D~-(b ){ ( ,2+ c2 ) -(b+)D1{ ( X2+c2)b+Db(x2+c2Ro ) } } .
Thus the solution of d2u 3 dt2 1-Ut2)2 is U=k(j l+)t2+k.t The principle illustrated in this section may be further exemplified and usefully applied , by attempting the solution of the general equation of the second order , qx .
D2u+ x.Du +x.u=P .
By the interchange of symbols , we have ( p(D)(x2u ) -4 ( D ) ( xu ) + ?
z(D)u =P or x2p(D)u + 2xp'(D)u+ pi'(D ) u -P -(D ) u -'(D ) u= P. + ' ?
( D)uJ This equation is soluble if Z(D)u= { 4'(D)-p"(D)}u ; for it then assumes the form xp(D)u+ ( 2p(D ) -4 , ( D))u==v-'P ; the solution of which , by ( 4 .
) , is u p(D)E/ D ) X-1(p(D))-2E4d f lp}D }x1P and , by restoring the symbols , we get for the solution of pLr .
Dlu+ lx .
Du+ ( 4'xp"x)u =P , u= _Wd{ ( ix ) 2fdxD(p ) } the correctness of which may be ascertained by verification .
If qAx=(n+1)p'x , we have for the solution of px .
D2u+ ( n +1 )px .
Du +np"x.u= P , u ( ?
x ) Jtpx)n-jPdxdx .
If 4x be made equal to Qpx , and P to Rpx , we get for the solution of D2u+QDu ?
Q'+QR , u= ~x .
&amp; -QxD ' { ( r ) 2 &amp; Q]D'(Rpx ) } .
And if px be made equal to SfT or ST , , we have for the solution of D2u+QDu+ ( Q'+QT-T2T')u R , U-S T , -QjD { gQl WlD ( R ST ) } , a form easily arrived at by ordinary processes .
IV .
Application to Partial Differential Equations .
Most of the soluble forms above deduced are readily convertible into analogous forms of partial differential equations , by substituting for the constants any function of Y , where D ' denotes differentiation with regard to a new independent variable .
TIbus if in ( 9 .
) for Q we write f(x , D ' ) and for c we write V-IkD ' , we have the symbolical solution of d2u I dI du d2u_ dP+2fx 7 kd2+ f2+fl M( ; _'))U=P , where P may be a function of x and y..f dnd For example , if fZv , d } be of the form y the equation becomes dqu 2n dOu Wn XdX n du m(m-1 ) d~d ~t ; pk2 } -FU 2= , A(w ) wz-~i X -d4 P die die of which the solution is u= eS ng -D(D2-k2D'2)m { Xl(D 2-_k2D 12)-1mj-(r4-1 ) nlogx.l qkxJ ( ) Now ( nox .
D`4(x , zi ) =4({ , y+n log x ) , and a-n1ogx.w kX , y ) ( y-n log x ) .
The question , therefore , is reduced to the solution of du d2u v75 -kdeY , T(x , y ) , or D2u-a2u=T(x , y ) , writing a for kD ' ; whence U=-2a QD S ` Y(2a Y , __ k N'(xy-k k dx _1 -kvDT ( x -)dx _2kD6 fx ( ..2kD1 jfN(Y ) which is k(N'1(x , )y)+T2(X , y))+X &amp; ( Y+kx)+p(y-kx ) , where X and pi are arbitrary , and NT1 and T2 are derived from IF as follows ; for ' in T write y integrate to xy+kx , and integrate to y ; this gives T from which '2 is formed by changing the sign of k. / d.d Iffy ( , g ) be of the form 4x.-+pX , the equation becomes +PU du2 d du / ~+2XX~d +((4X)2-k2 ) - ?
~d+~I+ px~+ U\ , ) ~xnm1J and the solution is u=xmi_-1ix6D'(D2-k2D12)m-1 { xI(D2 _ k2Y2)-mpw-(M-1)6jx4( y+( X ) } } , where S-XjxD4+(x , y ) means '4i(x , y-X1x ) .
Transforming this equation to new ' independent valuables p and q , by the equations p=y-X1x+kx , _q=y-_lX--kx , or 2k 'Y=X(2)+ we have , as a soluble form , d 1u np +qX Iddu &amp; f1 dF do , mm- ) dpdq ?
2J"( 2k J( td ?
~V)J + ( k A &lt; +2k(+ ) mqrnI ) ) u- ?
p(p q ) the function X disappearing in the process .
If px be of the form we obtain a well-known equation , solved by EULER in a series when there is no second term , dtu a Idu due a(a-1)-m(m-1 ) dpdq+ptq ( T+ ) ( ap + mWm-y which is therefore soluble ; the solution being ema(D2_D12)m 1 { X ( D2 -DI2)-m{X a-r+lp ( x y ) } } , where p1 is determined from p by the equations p w+y and q=x-y .
If a=O , we have the solution of d2u m(m-1 ) dpdqp q ; and if a=m , we have the solution of d$u m ( du du ) dpjq +pq adp + dq ) from which the solution of d2 4m d2u may be obtained by making s=0 S(paq ) -2t 2m-1 t=c(P-0 ) Returning to the ' general form and making Xx=0 .
we havre f~or the solution of d2Xu u +2i.du + p2 p m(m1 ) dy2 d2lmx XIs1 .
( X _ , r } du m2 Again , equating the factors of and u to zero , we haves= and 7xix2m ; therefore d+2xu d2u + ( xd4 u k2 d2u 2m du dx2 dxdy+ is soluble ; and if the variables be changed by the equations P=kx+2m .
+r(2ml ) Fie q= k , ,vx tq-l ) 2m-1 the form dpdy +2 2p+ k4(dv +P d8 ) becomes soluble .
By a process of a similar nature applied to ( 10 .
) , it will be found that the form ( ad1u du dnu dnu ( an+ bn)- ?
+ ( a , x+b 1 ) +Oip ? . .
+ ( al b dy- ?
( ax + b0)df( a x ) has for its solution u=(a D ?
+anD'.D'+ ... +aD .
D n1+aoD mn Ya ( D-cD ?
)A(DB ... x { { Ah(D aff ) -(D PD ) ixy } } V. Connection with Definite -Integrals .
It is well known that many of the differential equations integrated by the above processes , and whose integrals are in some cases capable of an expression merely symbolical by reason of the number of operations to be performed being fractional , may be integrated generally , when there is no second term , by means of definite integrals .
Now with reference to most of the equations of this description here integrated , I have observed that the symbolical form above given is capable of being instantly ( and , as it were , mechanically ) converted into a definite integral of the form eb UfMy Z.stfdz , a the function pz being typified in the symbolical solution by the form of the operations preceding the factor x-l .
To explain this , let us take the equation 2rn D2u+-Du-c2u0 its symbolical solution is au -k(D2C2)mfi { Z7'(JY ; c2 &gt; -m0 } ; and the assertion is that a solution of the equation in the form of a definite integral is obtained by writing for fz , k(Z2-c2)m-l and selecting the limits properly ; in fact , it is known that u= kf(z2 -c2)m lzdz is a partial solution ; and as it is a known theorem , that if u=ul solve D2U+ l+ IDu-C2U 0x it is also solved by u=mul if for n be written -n , we have for the complete solution U=k Z( !
2c2)m-lezxdz rkx-2m2 ?
( Z2c2)-Mrzxdz =k 2 1)m-dlsz k/ dz + kx -2mrn ( 21 )nmsdz .
1 d~~~~~~~~2U If for x be written ( ( 1-2m)t)-2mn-I we obtain the solution of dt2-a2tu=O , rm and C being properly taken in terms of a and n. In like manner , if we apply this mode of conversion to the umiore general form D2U+2QDu+ ( Q2+Q _ C2De )u O. of which the solution expressed symbolically is ux xm-Qi(D2-C2)m-lfx-l(D2-c2)-lO } , the definite integral ought to be u kxm &amp; -VQ:j~(Z2 1 )sxdz + klv-m+ I-Qf ( Z2 1 ) -msczdz , and this is in fact the solution .
The limits must be determined by verifying the equation and assigning them .
accordingly ; the verification at the same time establishing in the particular cases the correctness of the results arrived at by the substitution .
Thus , if we take the general form xp(D)u+ ?
( D)u=O , and its sy mbolical solution U= { &lt; ; ( D ) ; SX(D)~ , t , IS-(D)o } where zt==ftdt , the conversion here indicated gives u= kf Qz ) -la5ztzdz .
To verify this , we have p(D)u=fz Vx d 4'(D)u= / ' s dz ; f7 d( ?
Z ) =sz fsXdz , and the equation is verified , if ssox vanishes between the limits .
Whether the limits can be so taken depends upon the form of zz .
Apply this to xD'u+u=O .
Here 4:=znz Z-2i+I and u=k___--f+izdz .
z-n + The limits 0 and ac cause sn al to vanish and satisfy the equation .
So that f0 z-nf I UkJ Z s-%+i* dz is a partial solution ; which may be completed by writing for z , z , f3z , &amp;c. where 1 , ,f &amp;c. are the roots r i. Again , if we apply the above solution to In-7u=O0 , we have xz=- , 4 , z= _ zn~l and Umt/ -sdz .
n+1 Here , there do not appear to be any limits which will make r Th vanish ; but if we take for the limits 0 and oc we have Dnu-x-u=k .
This last equation is also solved / , o z_+l by u=kc J0 s+1 dz , a being a root of rt =i .
Therefore the original equation is solved by oc zn+1 U= Of gne ( -06 ) dz , which may be completed by the use of the other roots* .
Let us apply this method to the solution of ( anX+bn)D u+( .
( Mx+bn-1)Dn-'u+..+ ( al +b1)Du+(aox+bo)u=0 , and we have at once U kf(anZ +anz-1 +..+aiz+ao)(.(z MA(z_..3)B ... . ?
-dz , where bnzn +bzn-'+. .
bzn ?
bZn-l'+ ... bAB anZ +an-1z '+ ... an(z-a)(Z-P3 ) an Z+ a z* See MoI u 's Calcul .
vol. i. p. 644 .
To verify this and find the limits , we have x(a.Dnu+an , -D 'u+..+ adieu +aou)= k*jlZ )A(z bit -kS Y(zM)A( _Z)B SZkf#_*z M)A( ~t)B(2 !
i+ +BA zidz =(bzD u+ -1D8 + ... +b- , D +b)A(Z_ ++ San be~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~b , if kst(z-OG)A(z-OB3 ) ?
'can be made to vanish between the limits ; and this condition is satisfied if X be taken as the lower limit , and a , f3 , &amp;c. be taken successively as the upper limit , whence the complete solution .
If the expression TV+. .
has m roots equal to a , the form of the solution will be modified .
If in such case the rational fraction bzn +..bnA A A___ B_ a~z*+ ... a_ ++A ++ Z ?
+.*at ( a _m ( Z _C)m-l *+~tzf3 .
the solution becomes Am I Am-i 1 2A ' 'Mkjin " .
)1 e(Z-W ) m-Z ( .Z-a 2 ( Z UAl(Z -13)B a 6--xdz between the limits -oc and a , A , &amp;c. This solution is incomplete ; but it may be completed by using instead of adz ?
... its first , second , ... .
( nt-l)th differential coefficients .
There can be no doubt that this remarkable connection between the symbolical solution and the solution by definite integrals is not merely accidental , but is founded upon a similarity in the processes by which they would be respectively arrived at in a general system of solution .
The following considerations are offered as in some measure explanatory of the connection above adverted to .
The equation to be solved is of the fortm xp(D)u+iJ(D)u=O .
Now if u=kf w.dz , we have b ~~k k d('Vz-z p ( D ) uk =O as.pz sdz -ks1 , zp 6Z kg ?
z rdzas between the limits .
And or~ ~~~~ ~~~~~~o or Vz =Pz ( &lt; &gt; Zs , being the same process as that by which , in the symbolical solution , the form of the function of D preceding the factor x-1 is obtained .
VI .
Linear Equations in Finite Differences .
If in ( 3 .
) we change the form of the functions by writing a ' for D , we have xP(ED)U ?
+ D()U=X , or x.(E)ux+(.).ux.= Q*.(11 .
) where E is equivalent to I+ A , or denotes the operation which is performed in changing u , into u It follows that the symbolical solution of ( 1 1 .
) is u= ( 9E)E-1xEt1E-EQwhere Xdt:j~vdv if J= v. A case obviously interpretable is that in which 4vmnvw'v ; and in that case the solution of x(E)ux+mEp'(E)ux= Q is =-p(E)jm-1-1{p(E)j-mQ If we take the equation ( a x+ bn ) t++ ... .+ ( a'+bblu+1+ ( aO x+bo)u =Q~ we have v VfQ 1 +v_ , + ) dv , where a , , &amp;c. are the roots of anvn+..+alv+ao=O .
andu = ( anE + 1+aE ao ) -Eao ( E l(E-3 2jx**(E ao(E)Al(E-1 3)-A2 ... .
Q ) .
In like manner , if ( a x ?
bn)z"8ux+. .
+ ( a. , x+ b1 ) Aux+ ( aox + b0)zy= Z we have ub = ( an A ? . .
+ al + ao ) ' ( 1 + A)(A e_)AI(A _)A2 ... { X((I + A)-B(A _)-Al( AP Q ) } where vn'+..+b ?
v+b _B A_ A ( 1+v)(avn +..+alv+ao)1 ?
v V-+ + Thus the solution of x(E2 c2)u 2mEux=QzQ 2m 2 Qx or u , +2u x+ rt -c uz- , in sr ( E( m ) and that of X(zA2C2)u -2mAux=Q ol ' A2U _2mAU C2U X 2m 'n mf 2m mi ?
n iS Ux ( A2 ( c2 ) -E 2_+( A+ C ) o(-c)T(+c ) { x(Ei ( c ) ( 4 c)+Q ) } The results thus obtained are not interpretable unless B , Al , A2 , &amp;c. are integer ; in the event of any of them being fractional , however , the solution of the equation , ( the second side being suppressed , ) may be found as before in the form of a definite 9~~~~~~~~~~~~~~~~~~~~ integral , by introducing the factor asx and changing D into z , and writing for the integration with regard to z between the proper limits .
Thus it will be found that the solution of ( ax+ ?
bn)u , +r+ *. .
-(ajx+ bj)uj+1 4(a0x +b06)u=o is au , =Jcan ' A +..+ajl +ao ) ( -)A( )A ... .)ZX dg between proper limits .
For taking this value of up , , we have -~~~~~~6 ( S )A S x(an'o , ,n..+ alux+ + a.Z)-i ( o )A ( A_ ) ... S ) X ddz bo Z()( -+saG ( e &amp; )A , ( SZ_P)A2 .
d8ZX SZXS aO ( x )A(z_ ) -jb ?
ao ?
_M)AI ( ?
Z1 A2 ) ... ~2 ) Of these two terms , the first vanishes if one limit be z- , and the other have the successive values log a , log P3 , &amp;c. , and the second term is equal to -T ( aO(( ?
Z)Al(SZ_ p3)A2 b.)nZ ?
+b d= + od2 ( + .
+ b~U+ , + ba , ) , so that the equation is verified .
The complete solution therefore is 'log f LO UX C , _a , ( a~snz+.+a1Sz a0)Sao ( 6z _ a !
) Al ( Sz A,2.A..zdz Iblog on bo + C2 , / -0 ( anEn + **aS a0 ?
'0 ( &amp; -C)1 ?
j2 ... d + &amp;c. or more simply , Ux= CJ1 ( anV +..+av + a ) Ava(v ?
)A1(v 3)A2 v'Ldv f ~~~~~~~~~~~bo +c2Jo ( avn +.+avv+ao)1 v -j,0(V _)A1(V Aj..f3)A.Vx-dV + &amp;c. &amp;c. MIVcccXLvIII .
E In a manner in all respects analogous , it may be shown that the integral of ( ax + b. ) A"'u , S , .
+ ( adze+ bl ) Aux+ ( aox + bo ) ux= 0 , is ux =C 1 .
, av'+..+ajv+a0)'v( o)Ivp)~.( +v ) 'dv Ab is ~ ~~ u=cl 2( any"'J + alv + aO ) 1vao ( v _ )A , ( Up )A2 ... 1+V)xldV + &amp;c. &amp;c. If the expressions v(acnv+..+alv+ao ) , ( 1+v)(anvn+..+alv+ao ) should have two B or more equal roots , we shall obtain factors of the form S(V-_Y ) , as in the case of linear differential equations of analogous forms .
The process of changing the symbols may be used to obtain solutions of differential equations from known solutions of equations in finite differences .
The solution of the general equation of the first order ( 1 +A)ux-PxuS==QX is uQ iS tb ~~~~~=.J ... .
Pw I-PI(..Pn ... Pxr-2p Pa A similarity existing between this form and the solution of linear differential equations of the first order will be seen by writing the above equations in the following form , DU-X.u-X , and the conversion of symbols would give E log1 D1 -a- .
1 D+1 ) a symbolical solution apparently incapable of rational interpretation , at least in finite ternis .
If however we suppress X , and by a conversion similar to the one before proposed , pD be changed into p(-z ) ; and x be changed into -D ' , D ' denoting differentiation with regard to z and a factor s-- " be introduced , we shall find that the substitution for e1 is 1 , ; and the general result is where N denotes summation with reference to z , and the first summation must be taken between proper limits .
The form of this is between proper limits ; which , changing the sign of z , may be more conveniently written To verify this , we have S u= &gt; ( ... z 2).p(zl).s(zx ) ~P__suppose , p ( D ) u= I( ... .p(z 2 ) .
p(z I ) .
pz .
s ) =P , ; consequently e2uu-p(D)u=:E ( P-Pz ) = AP'P_=yx( ... *p(z -2 ) &gt; p(z1 ) , ) the last term being taken between proper limits .
Let a , P , &amp;c. be the roots of pz=O ; then if at one limit z have the values 1 +ca , 1+i3 , &amp;c. successively , P,1 will vzanish ; and z= oc will give a vanishing value at the Gther limit .
The form of the function zx must be ascertained by verification .
Let = z+azP(a)(z(z p(n ) ... .
p(z-2).p(z-l)={(n-e ) ... .(z I-ez)}{(n --i ) ... ( z-P-f ) } ... .
u= &gt; l'ffQp(n ) ... O(z 2).cp(z -1 ) )= .
n-~~n ) =P-( ( c)F(z i3).z ) = Pn.-c.)F~ -j ) dvsv1v1-af'-:dv2~ws 43jdv. .
-r(na ) r ( n_ .
o 8l-rrlr 2 ?
v3..Szx .
Now Ev1 2 -v.2 1 5x V2 ... VV2 .
-1 ' which taken between the limits -ox and 1 +a is a'lv 12..^1 and , if this be integrated successively with respect to v1 V2 ... and taken between the proper limits , we get +[2 ] [ a-Pf+2 ] Et -1y+2]..6n+ } If n=+ 1 , we have ( ocy+ 2). .
2+ ... } To verify this , we have S-xu 6 l+ ( a -+1 ) 9+ 1). .
e+ } p(D)u= - &amp; ma { ( a p+1 ) ( a e+ 1).{+ .
_x u_P(D)u=-s6 If all the roots of pz are zero , we have for a partial solution of s-$u Dnu = c , u-cx +v2+. .
+ ... . .
} dv -(VI+V2+-Vn ) =-cg~ ( 1+ ?
+ &amp; + ( 1x 2)I 2x+(1 2.3)n ?
3x+. .
)I Miscellaneous Forms .
In the course of the preceding investigations , I was led to attempt the solution of some forms of equations by means of successive operations not consisting .
exclusively of D combined with constants , but involving also functions of X. The only important result which I obtained is the following , being a slight generalization of the method originally employed by me in effecting the solution of the equation of LAPLACE'S coefficients .
' The equation D2u+ bDu+c2un(n+ I ) u-X is solved as follows .
Let a and j be the roots of z2+bz+c2=O , and assume Duau=ui .
Then Du1-fPul n(n+ 1)-ou-=X , Cos XU=n(n ?
+ )(Du I-43ul-X ) , cos2 X 2cos .
sinx ( COS2 X , Du-au= ) ( D u1 + bDu , + cu+ ) ( Du,43u , ) ( Dn(+ X =U or D2u ?
+bDul +c2u1 -2 tan x.(Du1l-6ul ) xu+)1 =X'"X-2 tanx .
X-Xsuppose .
Assume Dulau,2 tan x.u1 u2 , then by a similar process we obtain D2u2 + bDu2 + c2u24tanx.(Du2-j3u2 ) -n(no ?
u2=l ) X'- &amp; X-4tanx .
X , =X , ,suppose .
Similarly , if we assume Du2- &amp; 'u2-4 tan x.u2=u3 , we obtain D2u3+bDu3+c2u3-6 tan x.(Du3 f332 ) U3= ?
X ' , ax X suppose .
Proceed in like manner until we arrive at the assumption Do .
au .
2n tan x.u._ u.+ .
Then : D2 ?
1+ bDu+1+c2u , + ?
1-2(n+1 ) tan x.(Du.+ , -f3u+ , l)=X( , + , ) .
Let Du _PUn+i=Q then DQ-ocQ 2(n+ 1 ) tan x.Q=X(.+1 , and Q=Set(cos X 2(n+ ax ( Cos X)2(ff'lX(n+l)dwx u , ,l = tfe$)f(cos x ) s(n+l ?
/-an(cos 1 ) + Un-= e(cos *v ) s -f2(cos x)22uI +v1dx ; and so on down to u , which will be found to be , d -n ?
fim( ta ) J(A9)x ( cos x ) 2:fa(M- ?
X( cos x ) -2(n+1'fa~Z xV(coS X ) 2(n+t)X x(ns+ ldx2dxrdx- ; and if X-O , this becomes ( d u= ?
ks ( tan x ) ) fa(g3- &amp; )x(cos x)2:fS(i- ?
Z(cos x)-2( , + ' )dvdx } , proper precautions being taken with regard to the introduction of constants .
Perhaps the difficulties relating to the constants may be evaded by writing the solution in the form u =kex d { ( { G-z)xcos ( s)2(2n+1/ ( ?
A((CoSx ) 2(n ?
dx } , and then substituting ( n+ 1 ) for n , which does not alter the original equation , we have U= khan ' ( d(t a ) { 6(A- ' ) ( cos xV ) -2fS(a-)(cos xS)2nddv } .
If a and 3 are both zero , we have for the solution of D2u-n(n+ 1 ) u =O ?
u=k(d(t { a ) ) ( coSTx ) -2f(cos x)2ndx } =kQd)f { ( +y2 ) } ( d)( where y= tan x. Let c=c and 3=-c , so that the original equation becomes d2u c2u-n(n+1 ) X= ?
then the solution is u= kcx ( d i ) { f-2cx(cos X)2:S2cx(cos 2 )-2(n+ )dxdx } k*icx(d(I + ) ) { f -2c(cos x ) -2(X+ l 2cx(COS X)2ddx } , which contains the proper number of constants ; as the constant which enters by reason of the first integration disappears by the subsequent differentiations .
This solution will apply to LAPLACE'S equation , if for c be written cady This gives for the solution of d2u d2u u C2dn(n+ 1 &gt; 4/ d n\ U= S CXdY 6{ 2xdy(COS X ) 2'f(cos 2Vc ) } J2c , )dx\d(tani x ) d the symbol s2c'ydenoting that , after the operations are performed , y-2cx must be d written in lieu of y ; and the symbol seedy denoting that , after the further operations are performed , y+cx must be written in lieu of y. The solution is simplified byconsidering the latter function alone as a partial solution , and completing the solution by changing the sign of c with a new arbitrary function .
Now if in LAPLACE'S equation j(1 2 ) U + ?
+(n+l)u=0 , we make x=tan-'(PV/ 1 ) , we obtain dw2u-dyp-n(n O__ .
The solution of LAPLACE'S equation , therefore , by this process assumes the form 8ldy ( 1 2 ) h(y-2 tan-('v/ 1 ) ) } + , -tan-1(PV=1)dW- ) { _p42 ) ' x( +2 tan(ok , / - ) } POSTSCRIPT.-Received March 16 , 1848 The followingbrief investigation is more general in its results than that developed in pages 50 and 51 .
By applying the fundamental theorem to the linear equation Uxt+n-50 , U , ; , -4A , and its solution ux=p(x-n).p(x2n).*.1 { ( Pxp(x-n)P(x 2n ) ... )-l+ , xj ( where the sign of summation has reference to x , Ax beingn , ) we obtain the equation M"u-p(D)u=-*x , ( 12 .
) and its symbolical solution u=p(D-n).p(D-2n ) ... { i1((p(D).p(D-n).p(D-2n ) ... )+x ) and by expanding the factor A and reducing , we obtain for the solution of ( 12 .
) the series ( E referring to p ) , u=:z6-pnx(pr(D).-p(D-n ) ... .
p(Dpn3))godx =1((P(D+pn)*p(D+(p1)n ) ... .
p(D))I{6 -P"ZXl If s=ky , and vi consist of powers of y , the above formula gives the solution in series of powers of y of the equation ( YpQ ) uq + yn= y. Several equations of this form solved by Mr. BOOLE'S general method , are given in the Philosophical Transactions for 1844 , pp. 236-240 .

