23(5 
IOWA ACADEMY OF SCIENCE Vol. XXIV, 1917 
sets of five wires each were used, each set being drawn from a 
single block of tungsten. 
The diameters were measured by means of a micrometer- 
microscope which was first thoroughly cleaned and oiled. The 
micrometer-screw was tested for uniformity of pitch by measur- 
ing the distance between two fine parallel ^cratches on a pol- 
ished brass plate along a scratch perpendicular to them. Fif- 
teen measurements were made with each quarter of the screw 
and -the following results were obtained: 
Part of Screw Used Distance in Revolu- 
tions of Screw 
Left end quarter 8.864 
Left center quarter 8.832 
Right center quarter 8.836 
Right end quarter 8.817 
Probable Error 
0.0050 
.0050 
.0043 
.0041 
The two outside quarters are undoubtedly more uniform than 
the data indicate since it was impossible to line up the cross- 
hairs accurately for the outside readings, owing to a slight blur- 
ring at the edge of the field of view. The center half of the 
screw only was used in measuring the diameters of the wires 
and this part is uniform far within the probable error of the 
measurements. 
The micrometer screw was then calibrated by measuring, us- 
ing the central part of the screw, six separate millimeter divi- 
sions of a standard meter, the different divisions being chosen 
at various places along the bar. Five measurements were made 
of each division and from these data the length of each division 
together with the probable error of a measurement was calcu- 
lated. The greatest probable error of a single measurement of 
any one of the six divisions was 0.0197 revolution of the microm- 
eter screw. The mean of the six separate sets of measurements 
was 21.510 revolutions per millimeter. Since thirty observations 
were taken the probable error of the mean is 0.0036 revolution. 
The wire whose diameter was to be found was held stretched 
between two round brass clamps, with triangular heads, mounted 
so that they could be rotated about their geometric axes, which 
were also the geometric axis of the wire. Two readings of the 
diameter at a given point were taken with the micrometer-micro- 
scope. The wire was then rotated through sixty degrees by 
means of the triangular heads of the clamps and two more read- 
ings were taken at the same point. After rotation through an- 
other sixty degrees two more readings of the diameter were made 
