|om»-] PORCELAIN AIR THERMOMETRY. 20.*> 
ind T zn of the same table. The curves rf and e are supplied by way of 
sxample, and are values of T zn obtained by beating a Fletcher injector 
[muffle form heated by power blast) to maximum intensity and then 
illowing it to cool. A comparison of curves a, b, c shows that from C00° 
:he heating takes place sufficiently slowly to make comparative meas- 
urements admissible. Throughout the stage of heating Tand T zn agree 
within 30°, usually much more closely, whereas Tand T hg differ by as 
much as 90°. The maximum temperature is not above 1,000°. After 
:he gas is shut off the furnace cools too rapidly for calibration work, 
he thermo-couple cooling at somewhat greater rates than the air ther- 
nometer. The curves d and e for the muffle furnace, the form of which 
s a thick-walled cylindrical box, exhibit rates of cooling very much less 
jhan the curves a, b, c — a desideratum which is much more fully realized 
n work with the revolving muffle furnace, for which the curves h, Jc, i 
ipply. Unfortunately, owing to accidents, the statistics of heating are 
mly imperfectly shown ; but beginning at about 900°, which is here 
learly the maximum intensity for the amount of gas supplied, Tand T zn 
liffer by only 10°, T and T n<J by about 80°. The curves then rise rapidly, 
>wing to the fact that the influx of gas has been increased, until at 1,250° 
rand T zn differ by nearly 40°, Tand T hg by nearly 150°. For reasons 
)f a practical kind, the temperature was not increased above 1,250°. 
3elow 1,000° cooling takes place at rates sufficiently retarded to make 
jalibration work .practicable. Above 1,000° it is expedient to obtain 
lie different degrees of constant temperature by regulating the supply 
)f gas. A comparison made between the curves Jc and h throughout the 
jonrse of cooling leads to the inference that the well-known thermo- 
electric formula e=ar-\-br(j is insufficient for interpolation. 
The extrapolated temperature T hg differs enormously at high temper- 
itures from its air-thermometer value, the latter being the greater. 
3ince below 900°, T< T m1 the discrepancy can not be referred to friction 
)f gas in the capillary tubes, the effect of which would be of the oppo- 
site sign. The only cause which would tend to cool the air thermometer 
it a greater rate than the thermo-couple is the effect of the entrance of 
:he air above the meniscus of the manometer, while the mercury is 
gradually moving upward from a lowered position into contact with the 
klucial mark. Or, finally, the error T—T Z)l may indicate superheating 
n the calibration work. 
It is by no means the object to furnish in this place more than a sta- 
tistical diagram, as it were, of the degree of accordance, which the high- 
:emperature measurements made in widely different ways, present. 
When comparisons are made in the Bunsen muffle the ascending val- 
les of T zn exceed the descending values of T zn for the same T. The 
iscending and descending values of T zn in case of comparisons made in 
;he revolving muffle are nearly the same. This is a pretty fair test for 
dentity of environment. Nevertheless, if all values of T m obtained be 
aid off as functions of T, the band or pathway thus obtained is in some 
(857) 
