
A Quartz-Thread Vertical Force Magnetograph. a9o: 
increased to S=1300° and there kept constant, the loss in 
weight was practically the same as in air at the same pressure. 
The disintegration increased rapidly as the pressure was 
reduced, until at 0°2 mm. He it was more than ten times as 
great as at 30 mm. The disintegrated particles in this case 
formed a brilliant metallic deposit on the copper wires in the 
tube, whereas when oxygen was present the deposits were 
always black. 
Platinum and iridium were also heated in the high hydro- 
gen vacuum, but showed no disintegration. A few observations. 
were made at lower temperatures where the disintegration 
first begins to appear. Iridium showed a loss of weight at 
the lowest temperature, but the results were irregular since 
the metal was oxidized at this temperature. Experiments 
with gold in oxygen were also attempted, but no traces of 
disintegration were observed. 
This investigation has a practical bearing on high-tempera~ 
ture work in general. Platinum heating-coils when brought 
too near the melting-point of the metal are rapidly destroyed, 
and the case is still worse with iridium, which otherwise on 
account of its high melting-point would be particularly suited 
for use in pr oducing high temperatures. Our results indicate. 
means by which the platinum metals may be protected from 
disintegration. 
Physikalisch-Technische Reichsanstalt, 
Charlottenburg, Feb. 1904. 
XLVI. A Quartz-Thread Vertical Force Magnetograph. By 
W. Watson, A.R.C.S., D.Sc., FRS., Assistant Professor 
of Physics at the Royal College of Science, London * 

[Plates XXII. & XXITIT.| 
HE late Dr. Eschenhagen was the first to show that 
when the moment of inertia of the suspended system in 
a horizontal-force ma gnetograph is made very much smaller 
than that employed in the ordinary form, it is possible to 
detect variations in the earth’s field of quick period. Since 
the study of these short-period variations promises to be of 
considerable interest, the design of self-recording instruments 
capable of giving satisfactory records becomes of importance. 
Both Dr. ‘Eschenhagen + and the author { have described 
* Communicated by the Phy sical Society : read February 26, 1904. 
+ © Terrestrial Magnetism, v. p. 59 (1900). 
{ Ibid. vi. p. 187 (1901). 
