280 8. W. Holman—Method for Calibrating Thermometers. 
. 1, and that of the upper end u,. Move the thread less 
than 1 mm. and read again, finding thu sl, and w,. e the 
thread about lem. and r ae l,and u,. Move the thread less 
than 1mm. and read /, and u,. So continue throughout the 
whole length of eden increasing the number of settings 
or repeating the whole series in reverse order and several 
times if the highest attainable precision is desired. This alterna- 
tion between 1 mm. and 1 cm. in setting tends towards the bet- 
ter elimination of errorsin estimation. It is not, however, essen- 
tial, nor even always as well as an equal number of distributed 
readings. This must depend upon the skill of the observer. 
Avoid, as far as convenient, taking readings with an end o 
the thread. apparently just at the line of the scale, as the 
width of the line, even in the best scales, is a source of con- 
_ Then u,—J,, u,—l,, etc., will give a series of lengths of the 
calibrating thread in all parts of the tube. Before reuniting 
this thread to the rest of the mercury, plot points with abscissas 
» 4, ete., and ordinates u,—1,, u,—l,, ete., the corresponding 
lengths of thread, and draw a smooth curve through the points 
of the capillar e; and, sho any parts of it show con- 
siderable irregularities, the wah thee LSS totes of the tube 
should at once be re- a! ored with the 
f the sucabonaci is aioe ed one or two 
starting point. 
upon the curve the ordinate w’ corresponding to the 
abscissa A; then with abscissa A+w’ find the corresponding 
ae ws 3 with cpm A+u’+w’, find the corresponding 
ordin ’’, conti o the upper limit of the graduation. If 
A is on a edidient Beau from the lower end of the gradua- 
tion, find a similar series below the point A. These points, 
A, A+u’, A+u’+uw’’, etc., upon the graduation are separa 
by equal volumes of the ‘capillary. Select any one of these 
as the second point of which the error is to be arbitrarily 
* Some of the advantages of Neumann’s method are offset by this error, 
