XV .
Of the construction of Logarithmic Tables .
By Thomas Knight , Esq. Communicated by Taylor Comb , Esq. Sec. R. S. Read February 27 , i8i7 .
1 .
1 HAVE endeavoured , in this short Paper , to give a simple and connected theory of the construction of logarithms , which I think has not hitherto been done .
PRoP .
I. To find the Logarithm of i+ x.* It is not difficult to see that we may assume L(i4 ) 'Ax + " Ax ' + " 1Ax3 + " " Ax + &amp;c. , whence , L ( i + y ) -'Ay + / 'Ay3 + " 'Ay3 4 " " 'Ay4 + &amp;c. , and L { ( i+x ) ( '+y)}-L(i+4x+y+xy ) , or putting 1+x_=r , { ( + )y_'A(x+ , ry)+t'A(x5+ , ry)9 , " A(x+.7ry)3+ &amp; If we substitute these three expansions in the equation L(i +x ) L(i y)-L{ ( t1 x ) ( a y ) } which expresses the nature of logarithms , and compare the coefficients of the first power of y , we find 'A = 'Ar + 2"Arv + 3S"AirH + 4 " " Are 4 .
&amp;c. or 'A = 'A + 2"Ax + 3"/ Ax2 4 .
4"11`Aa3 + &amp;c. + 'Al +42"AJ + 3"Aj whence , by comparing the coefficients of the powers of A , 'A='A , 2 , " A4'A=o , c3"'A+4 .
1A o , 4"/ A+43.'A=o , &amp;c. ; or 'A ~~2A'A 3A/ IAA__A A_ 'A A=-Wz , tA ) " 'A__2 A= A OPERA_ 3A &amp;c. 2 ~~3 3 ' 4 and L(ic , ='A { § 34+I2 ~3 1 As for 'A it may be evidently taken at pleasure ; and innumerable systems of logarithms may be formed by assigning different values to it , for 'AL(i+x)+'AL( , i+y)='AL(@+xI)( ily ) } which expresses that , if every logarithm in a system be multiplied by the same constant quantity , the products will -still form a system of logarithms to the same numbers .
Cor. By an easy transformation of L ( i + x ) , we get for BRIGG'S logarithms , M being the modulus , aZ m a-b I ab3 I ad-b 5 L-M +-(S)l4.(I)+ } and when .
o b-M a+b +3 Cib ] +5 ta+bl 3tadwe ever the logarithm of a fraction is spoken of in the following proposition , it is supposed to be found by this series .
2 .
How are we to begin , in forming a table of logarithms ?
DELAMBRE ( Preface to BORDA , P. 75 ) says , that we should begin at ioooo ; and the same writer ( Memoires de l'lnstitut , Tome cinq .
p. 65 ) , speaking of the great French Tables , says that the logarithms of primes under iomooo were calculated directly by series , and those of numbers above 1oooo by six orders of differences .
Now it is not easy to see , why any of the logarithms in the lower half of the Table , except those of the numbers 2 and 3 , should be computed directly ; since they may be got , each by a single subtraction , from those in the upper half .
Suppose , for instance , there had been found directly the logarithms of numbers from ioo , ooo to 200,000 ; those of numbers down to 3oooo are found by merely subtracting the logoa rithm of 2 , successively , from those of all the even numbers ; beginning at the topof theTable , withLi .
9998 , L 1999996 , &amp;c. and setting down the remainder for the logarithms of the ' successive numbers below tooooo , viz. L. 9999g , L. 99998 , &amp;c. When we have got down to oooo , if we were to proceed in the same way , we should have to operate on the logarithms thus obtained , between iooooo and 50,000 : If , therefore , we fear any , accumulation of errors , we may ( because 3x49999 149997 ) subtract the logarithm of q from L. 149997 , and from the logarithm of every third number going downward , and set the remainders down successively for the logarithms of numbers below 5oooo .
And thus we may proceed till we get somewhat below 3 , ooo ; then the logarithm of 4 Will carry us down to 50 , ooo ; and the logarithm of to 20oooo , which completes the work , those below 20,000 having been already found , In the great French Tables , however , it has been thought proper to calculate the logarithms of numbers under io , ooo with more decimal places than the rest .
These must necessarily be found independently of the others ; as they form in reality a separate Table .
In the next proposition , Is contained a general method of finding converging series for the calculation of logarithms .
The propositions which follow this are only corollaries from it , and give forms for interpolation ; so that every thing relate .
ing to the construction of logarithms is effected by one simple and uniform process .
PROP .
II. .
To express a number ( x ) by the product of a series of fractions con erging continually towards unity .
Let n , at , it " , &amp;c. be numbers much less than x , in the equation x- , W change a , in the second member , into + n , and mnultiply by such a factor as will restore the equality ; there arises ( x + n ) x If , in the second member x+ n of this equation , we change xv into x+ ni ' , in the last factor Z and multiply by such a fractional factor as shall again restore the equality , we have x+n x( ?
+n'v ) I ( X +n ) x-+n-J n ( X+n)( v+n If here , in like manner , we change x into x+n " in the last factor , and restore the equality as before , by annexing a new factor , then x+W ( x+n ) ( xin+d~n " ) =( , +n ) x Ax + X(x++nn ) x+n+n*)( +X+f ) ( a+n(x +t )( nf( + no Jr i+ mynk ( x+n)(x+n')x+n0)(x+n+'+ie ) and the same process may be repeated as long as we think it necessary .
Now it is plain that the last annexed factor , as we continue these operations , must always approach nearer to unity than that which was the last before ; thus , n being very msall compared with x , does not much differ from unity , and when x+nt is put for x in this fraction ( at being also very small compared with x ) its value will be nearly the same as before : of course the annexed factor ( X+ r +nlg will differ very little from unity : and it will differ from it SAl much less than the preceding factor for let = ie-U , i.-2 , and pt and [ L being small fractions ; the new fatrx(x+n+n , ) I-P factor I ( +n)(x+7')-[-b,1 -(p , -fZ ) nearly : and consequently differs less from unity than the factor which was last before .
4 .
These equations , being put into logarithms , give a series of converging expressions for the logarithm of x. We have successively , ist .
La.t=L(x+n)+L( )d L.L.-L(x +f ) + L(x + n)-L(X +n +n+ ' ) +L X10+0 xt +(n+n')xznnI but before we put the third equation into logarithms , it will be better to simplify it ; one of the most obvious ways of doing which is to make n+n'-n " then 3d .
L.x=L(x+n)+L(+n .
' ) +L x+ 2 " g ... L(x+n+n " ) -L(x+nt+n9 +L ( 5+3n !
~x1+(nnl+2nftUx xs +3nie+(nan ' +2n'* ) This may be still farther simplified by making n=n ' , consequently nfl"2fl , then L.x-2L.(x+ n &gt; )-2L(x+3n ) +L(X+4n ) x3 + 6nxl + gn~x + 4n3 If now we change n into -,1 and x into x+2 , we shall fall upon the elegant formula of Mr. BORDA.* In like manner we might investigate approximations of the fourth and following orders : but this kind of research has very little use , and the Proposition was inserted for a different purpose .
PROP .
III .
5 .
SupPosing that , in the Last Proposition-n'cn"1 &amp;c. =-1 .
It is required to find the law of the converging expressions for L x. In this case the four first transformations give x=.(x-i ) XA w='(w-1)~X ( - .
, V= ( Xi C(x-I(-)(--)(- ) ~- ) x. .
JS-I X 1X X_ 3 ) LI ( x x ) ( X3 ) T=2)1 L. =4L(X- ) )-6L(XX2 X X2 I ( I ( X3 ) ( x(1 ) ( x3 -0 X(-2 ) ' ( X- .
) X=(X ) Y. 2X( X_ -X which put sinto logarithmsd give L. x=LJ(x-i)tL X_ x(x-2)IL .
Lx--3 ) +L ( X_ - ) 3 ( _.3t L~x..-4L(x-i)-6L(x-.2)+4L(x-- !
, Q )-L(x-4 , )+Lx(x":z)'=g wnhere a coincidence may be observed between the coefficients and those in the binomial theorem ; and it is easily shown.that the same coincidence will have place , bow far soevrer ye continue the method ; or that , in general , the converging expression will be n(nI nln-I )(n-2 ) ( n-3 ) X XX ( X_2- ) 2X ( X 4 ) 1.2 34+ &amp; C. ( a{ 1.2.3 4CJ -':"3 For , if it be denied , let this represent a single result , to pass on to the next , we change x into x1 , in the logarithm of the fraction , and add a new logarithm ( L ) to restore the equality : the equation will thus become T x_ n1 tn-ifl _ _n(n i ) ( n-z)_ !
I ... I 1.2.2 4(-2 FLxL(x-ix-2 &gt; L(x-)* &gt; +L or L. ~~~ ~~4x -3 ... +L +11 IIn ~)J+ ~~~~~~~~~~~1.2.3 or LX x=n+ l L(x 1 )(n + z)n L(x-2+ ( n+l ) n(n I i ) LwA+(L ) and , by transposition , we find r ( n+i)n ( n+fln(n-i)(n- ) L)=~L$~~2 1 .
1l.4 xn +I ( n+ i)n~n-i ) L ( X-I ) X(X-3 ) 1.2.3 X &amp;c. so that the whole expression is of the same form as before , which is therefore proved to be general .
6 .
, iCor .
i. If in the values of x. in the last article , we put for x , in the second x+1 , in the third X+2 , in the fourth av+I3 , and so on ; and moreover represent the last fractions arising after such substitution by a , ao a " ' , &amp;c. , we get the following set of equations X= ( X-1 ) Xca X +1= XX ae Xex +2= ( x+1)%i X+xlx ' ( b ) X+3 ( x+2 ) )t+2 x(x+2 ) 0 a , which , from the manner of their formation , are subject to this law , that the mth fraction ( provided it is not the last ) in the value of x+n , is equal to x+n-i divided by the product of the first m-i fractions in the expression of x+n-1 .
If , for brevity , we put L , LV , L ' , L " , &amp;c. for L(x ) -i ) L. x , L(x+l , L(x+2 ) , &amp;c. the last equations give L0 -L +L.o L ' LV +L .
&amp; +La ' ( c ) L " I =L +(L'-LI )+L .
a/ +L .
at " -L " =L " +(L"-L ' ) +-(L -2L ' 4U ) +L .
o"'+L .
cc " ' L= L"'+(L"'L")+ ( L"'2L'+ L')+(L"-3L"+3L-L+ ) + L. l"t'+L.ctc " ( it .
eO(+ )Ln +i ( I.fn | L"(n-J+Lf _2 Lo ( t2))+ ... .
0 ( n ) n(n- .
)L " ` ( 2 L " ... ( n- ' ) )+ + ( L nL(n+ ) These equations are subject to a law arising from that which we noticed in ( b ) , viz. that the Mth term ( provided it is not the last ) in the value of L " n is equal to Lt'-'( " ' ) , minus the sum of the first mi1 terms in the expression L " " ( n- .
By term I here mean the whole expression included between two brackets .
If we form Lca " generally from the last term of equation ( a ) we have ( n+i)n ( n+ i)n(n-)(n-2 ) , ( x+n)x(xin -z ) 1+3x(x+n-4 ) 1.234 X &amp; C. L. al ... eLn +1 ( n + l)n(n-1 ) &gt; -L ( xenon )1x X~+ n3 ) 123 X &amp; C. J If any one should not be satisfied that the form given to L". .
( n+ i ) is general , he has only to form L X. .
( n + 2 ) from it , and he will find the same form in that case .
Now L"(n4- ) is formed from L"'.(n+ ' ) by changing x into x+i ( or L " ... .r into L".(r+ ) in all the terms but the last L. a"I'v(n+ ' ) , and this .
we may remember had x first changed 'into x--i , and afterwards into x+1 , so that it receives no alteration .
Finally there is to be added L '..n z. 7 .
LCor .
2 .
Change x into x-1 in equations ( b ) , and let the new values of a all , a " , &amp;c. be AA~A , &amp;c. , and there arises the following series of equations , x-I=-(x-2 ) X)(p x =(X-i ) xxx X(X-2 ) XS41(.)XI )xs x 5 ' X &amp; C. &amp; C. which , being pultintoe loarths gqaive btefis yte an putting aaa aa ax sa'xax s a0 and so on , give X+ ' ILaaa(fl ')+(L " ' x ( n n(nI)x )L+..(n , + ) 4( L. .
n"** ) + nn--3)Lt* ( n-2 ) ni-xt--)L .
t'.(n_3 ) + +L..n 1 .
*1.~I2.3 ~)~+ L.s".'.(n+ ' ) ; any immediate term , as the m+ 1tb , will be of the form L " ... .n_(m-)L"..(n-)+mX(m-3)L8u(xz ) 1.2 1.2.3 We easily see that F ( n+Q)X(n-2 ) _n___n(n_Ox_ -6)_ ( x+n)x(x+n-.2 ) 1.2 X(X+n-4 ) 1-2.3.4 X &amp; C. L.s _(n+ i)n x ( n-4 ) ( x+n-i ) Qx(x+nn-3 ) 1.2.3 X &amp; C. It is scarcely necessary to say that L. cx " ... n , L. sll ... n are to be expanded by the common series for L. a. viz. b* These forms are analogous to an expression in the method of differences , which , though not noticed by ST IRL1NG and other writers on interpolation , may be useful on many occasions , as the coefficients are small and few in number .
BORDA 'S expression for logarithms is a particular case of it .
n x(n 3 ) n(n-i)x(n-5 ) n _n(n-i ) ( n-2)x(n-7 ) U_ )n-1 1.2 3n-2 + 1-2-3 n 1-2-3-4 -Un-4 + AnU + / AU_i .
If we make un L. x , we have , by taking is ( in the coefficients ) successively 1 , 2 , 3 , 4 , &amp;c. L.x-L ( X2)+ series , L. =L ( xi ) + L(x-z).-L(x_.3 ) + series , L* x=z L(x-i)-L(x-3 ) } +L(x-4 ) + series , ( BORDA'S if we change x into x+2 ) .
L a_ co~b +I ( a-b\ 1vbt &amp; L A=2M -a 3+5 + &amp;c. 3 though , for the , ~a+bm 3a+b , __ purpose of the two following propositions , they will be better when expanded into series of monomials ; see Prop. VI .
PROP .
IV .
8 .
To construct a Table of Logarithms by means of interpolation from the converging expressions L. a , L. a ' , L. a"l , &amp;c. When treating of the equations marked ( c ) , we noticed a law to which the terms are subject ; this law affords an easy method of eliminating the second , third , &amp;c. terms , and , bx , this means , we find , successively , L. x -Lrx-1 i)-L.o L(x+ i )-=L .
x +L.a+ L an ( e ) L(x+2)=L.(x+i)+L.a+2L.Us+ L.a ` L(x+3)=L(x+2)+L .
a+3L .
a+3L .
a/ '+L .
a " ' *~ ~~~~ ~~ ~. .
L(x+n)=L(x+n-.i)+L.a+nL.a'J+ n22L *a"+ If any one doubts whether this form is general , for every value of n , let it be only a single value ; and supposing a , a ' , a " , &amp;c. to become a , a , a " , &amp;c. , by the substitution of x+i for x , we have L(x+n+-i)=L(x+n)+L .
a+nL .
a'+ ' !
'2Lo cs"4 aI Now , if we consider the manner in which the last fractional factors , in the values of x , at the beginning of the last proposition , were formed from one another ; and the change which they afterwards underwent in forming equations ( b ) , we shall easily perceive that Z ? . !
.(n~~~~~~.(n i ( 7Y ' whence L. * +L a by means of which equation , L(x+n+ i. ) becomes L(x fn + 1 ) =L(x fn ) +L .
oc+n L. a6'+n(n si)L.6 ' n(ni ) n 2 ) Le o~~~~~~~~~~~~~~~t4_ ~~~~~~~~n La , ,/ + nl~ z }n.o , I.23 or L(x+n i)_L(x+ n ) + L. o+ +L .
, '+(n+ , nL .
I'+ ( nx)ntn-T ) L. c"'-+ the same form as before , which is there1I2.3 fore general .
9 .
By means of equations ( e ) the logarithm of a number is found from the logarithm of that which precedes it , and the process will consist of a number of additions , equal to the number of the values of x we make use of .
Thus , if L. " be so small , as not to affect the last figure , to which we intend to carry the logarithms , it may be neglected and we must make L.c+ L.x ' b. L. .
+2L .
L =r+L .
a"+ L. a " r ' L. x+3L .
x'+3L .
a"zr'+Lt +2L.aos"=r " L. oz+4L .
az'+6L +L ' .
a'+3L .
r " ' O~~~~~~~~~~~~ 0 L(x+2)-IL(x+1)+r +L.ca'+L.c " ==L(x+i)+r ' L(x+3 ) =L(x+2)+r ' +L.z'+2L.ac " =L(x+2)+r " L(x+4 ) =L(x+3 ) +r " +L .
a1+3L .
a " =L(x+3)+r " ' &amp;c. &amp;c. &amp;c. If , in the value of LX** ' n ( Art .
6 ) , we put successively 0 , 1 , 2 for n , we have L tx ) ( +)r Z)X3 L~~ .
_ItL -J^L1fL.o 1 ( X-1 ) ( X+ I0 or L. =Mt ( - ) + )+ } 2X-I 3 2Xt-I 5 2X-I1 t~t=-GMT2XZ3 2x ' I ) +5 ( 2Xt-I ) }1M{ zX+ II( 2X+I 3 L. 2M77_ + .
z IJ *M+(-1 ) .
t/ +(n-i ) ( n-2)L.o'_ r ... . .
( n -3 ) + I.C f ' ) n.0L a " ' = rall , r ( n.2 ( 1 ) If L. al " " had been used , we must have made L. e"+L .
a/ / ' r , and have proceeded as before .
( 7 ) The substitutions above being made equations ( e ) become L. x-LL(x- ) +L a ( Lrv+i)==L .
x +Lo+La ' -L .
x+r=L .
x+ : L(x+2)=L ( x+1)+r +L.'+L .
a " ' L(x+ )+ rl L( X+ i ) +r+ rQ ( 1 ) L(x+3)-L(x+2 ) +r ' +r +L .
a"+j L.al'=L(x+s)+ ( I ) r " =L(x+gz)+r'+ r ( ' ) L(x+4)=L(x+q ) +r " +r ' +L..z'+2L ez"'-L(x+3)+ ( ' ) r"'S=L( .
+ 3)X+r"+ r " ( .
) ( ' ) -~~~~~~~ r""-'L( x+ )+r"a'+r"'t ( ) &amp;c. &amp; C. Where it is plain that eachi logarithm is found by four additions r ' , r " , r " ' , &amp;c. being got by two each .
PROP .
V. 1 1 .
To construct a Table of Logarithms by means of interpolation from the converging expressions L. s , L. s ' , L. s " , &amp;c. If we consider the formation of equations ( d ) , we easily perceive that the terms of L(x+n ) observe the same law , with respect to those of L(x+n-i ) , which we observed in equations ( c ) ; we have therefore by a similar elimination L. x L(x-2 ) +L .
s L(x+i)=L(xi)+L.s+ Les ' L(x+ )=L .
x +L.s+2L.s'+ Ls " L(x+3)-L(x+l )+L .
s+3Lo s'+jL .
s +L b3 L(x+n ) = L(x+n-2)+L s+nL .
s'+"(n i L. s'+ which in order that the logarithms maybe got from one another by addition , must be transformed as in the last proposition by the assumption of p , p , pt , &amp;c. ; p p , p ( &amp; ) oi ) l &amp; C p p ' P , '1 &amp;c. : thus if the case only requires us to use L.s , Les ' ( )(1 ) ( l ) and L. s " , make L.s+ LPs'=P L.s+2L.s'+ L.s " p+L .
s'+ L. sP L.s+3Les'+3L.ss ' -p'+L.s'+2L~s " ap " &amp;c. &amp; C. &amp;c. by substituting which our equations become L. x L(x-2 ) +L .
s L(x+i)=L(x-i ) +L.s+ L.s ' =L(x-i)+p L(x+2)=L .
x +p+L.s'+ L.s " I st L.x +p ' L(x+3)=L(x+i )+ p+L s'+2L.s " =L(x+i )+p " &amp;c. &amp;c. &amp;c. If now we put successively o , i , e for n in the value of L. .
n , given in Art .
7 , we find l. s=L( x9 L -s'_L( + Z)( ) Las -L((+X:2)or L. * _ccv x+ H ( i &amp; C. _ L. s'-_ 2ML{ , +(_)+ s(-x + &amp; C. e--I I\3 , if x L.s'= _ M{ ... .
-+ ) ) '+ &amp; cj 3S 3 XX-X 5x 3-3x ( BORDA 's Series .
) PROP .
VI .
ie .
To expand L. e~ ' ... .n and L. s " ... .n into series of monomials of the form A- .
L( +n)=L .
+M ** ' ' n+1 72f tf n+r n-Xi l+i ( n-i ) ' _+ I L(x+n_-i_ IL .
x+M _x *2Z I t+ ( t I ) + ( n + iOn __________n ( + ~n(n2 ) L~~~x~~~n.~-L x+ +M ~e 112 1.Z~n-gJ= s2 1.2 tiZX 12 2X ( n i ) ( n2)r_ + 1.2 rr+7 TIhese added together will give L It is easy to see that L.* x will disappear , because its coefficient ( i. .
)fl+I ; we have then , putting X to represent the sum of the terms formed by the diflfrent values of r , 71+1 ( n+ J~n -(n+j)n~nI)( n-+M r. t gXy+( ?
f~_~_( : ( n1 ) ( -3 ) r+ | a~~~s=+ -IMr 1 1.2.,3 k ~~ ~~ ~~rxr f where for r we are to take every whole number from one upwards ; thus L. M= L 9,33 terms of L. a , with the sign changed .
L. i } as is expanded , by means of the logarithmic series , in a similar manner .
PROP .
VII .
13 .
To calculate a Table of Logarithmic sines , or cosines .
It is quite evident , that , if , in the fractional products , in Propositions II , and Ill , instead of x , X-X1 , -2 , &amp;c. we had used successively cos. , cos. ( x.u ) , cos. ( -28u ) , &amp;c. or sin. , sin. ( x-u sin. ( x-2u ) , &amp;c. , the reasonings made use of would have been equally applicable ; and that the whole methods given in Proposition III , IV , V , including the gene .
ral expressions for L. a ' ... n , Ls , $##M , ( but not the expansions of the said expressions ) will hold good here , after we have made the above mentioned substitutions , Thus if it is L. cos. x which we are calculating , e we shall have L. aLCOS .
xo*L c-IL ( os .
( x-u ) cos. ( +u ) La " 'kcos .
( x-u ) Cos. , X_ CO. ( X +zU ) cos.3z ) Cos. ( X-U , ) cos.3( +u ) These are the logarithms of numbers converging contie nually towards unity , and must be found by the form for Lr Ia

