Calibration. --Besides the .set to be calibrated there must be provided a
standard set, also a third set to be used as counterpoise. This latter set
may he of any cheap weights as the actual values of the pieces do not enter
into the calculations.

Begin with one of the l-gm pieces. Place a l-gm counterpoise on the
left pan and the corresponding standard piece on the right pan. <'arcfullv
lower the beam rests, and then the pan re*,st,s, and .start the balance swinging
by lowering the rider momentarily to the beam. Note the xero point and
adjust the. rider so that the pointer swing* about the true xero of t he lower
scale. Now, without moving the rider, raise the pan rent;-; at the moment
when the pointer is passing 7,cro, then the beam rc,st:-5. ^

Remove the standard piece ami substitute the piee»» ' '

of the set. to be calibrated. Itcpcat t he dct cnnina.- :

tion of /ero point. If the latter lm>; not been
rlmngcd by the substitution of weights, the standard
piece and the experimental piece have the .same,
value, irrespective of the value of the counterpoise-.
If the xero point ha-H changed, shift, the rider to
rcHtore equilibrium. The amount of shift given the
numerical difference between the two piece;-,. If
the shift of the rider is lo the, right the in pen-
itiental piece is liyhtrr than the standard piece and
if to the /*// it is htuntr. Apply the indicated / ^

correction.

Repeal tin* process j»«t described, comparing
each piece of the entire >>et with the correMponding
piece of the standard set, finully tubulating the
corrections. The*«* should l»e recorded on ?i card,
which may be placed In the. balance cawe for future

If ait iieeunitely ^liindjirdi/.ed complete xet. of
weights in not uvuilablf*, th<* set may be calibrated

to a Mingle stniitfanl piece, or simply relative vnlii4*H •!}|1||»r* .Ml
of the* vnriouH pieces may b«* e«tahjh}n'*|, UHIIIK the i«,t t»»Iitu
method of Kirhunb."'

ife:rtlpft

Volumetric Apparatus. It bus alrfwiy hr<*n .shown tlnit tlic
baluiiec* in roiiwrnwi, <Iir«»rtiy or in*Iin*rtlyf in nil deferniinnf\nnx
nm<h* hy voluiwtriV pr<wi^>*c»H. But tin* work in irefulially
(IifTc»n*!it front gnivifiiHrte iiniily^is in fhiit in* final wHfiltiftu: of
pn»<'ipitut«is is made, hint end fit' fluV a nH^u^tiri'tn^fif i^ inmli- of1
tin* volume of a stumiunl sohifitirt n*quir<*<l to rumph't*' it i|i*fmitc
reunion with the Hihsfnnre uii<li*r inv»'><tRation,
1 J, Am, Chit/i. AW., 22, 144 UIMKI/; MAHIN, "(^uautitnf»v» Vi«.ilv-:|-:<"
2nd ed,, «M» <»<».