596 
DR. J. T. BOTTOMLEY ON THERMAL 
globe shall hang at the centre of the spherical enclosure : the ends of the platinum and 
platinoid are then brought together and inserted into a hole drilled m a copper plug, 
and are silver-soldered to each other and into the hole. This copper plug is firmly- 
fitted by screwing into a hole, which is carefully cut in each of the cojDper globes. 
The operations just described are, of course, cariied out at the commencement of a 
series of experiments, the copper plug being once for all attached to the thermo- 
junction. The globes can be attached or detached at pleasure for the purpose of 
having the surfaces polished, sooted, or otherwise treated. When a globe has been 
hung in its place for experiment, the lower half of the spherical enclosure is brought 
up, pressed firmly to the upper half, and soldered there. 
It has already been mentioned that, during the cooling of the heated globe, the 
enclosure is immersed in a, bath of cold water, which is kept at constant temperature. 
An ordinary chemical water-bath of copper is used for this purpose. It is shown 
surrounding the enclosure aa. One of the thermo-junctions being, as has been 
described, at or near the centre of the cooling globe, the second junction is soldered to 
a small lump of copper, which is tied to the bulb of a thermometer and immersed in 
the water-bath. The platinum Avire is made sufficiently long to pass from the globe 
up through the brass tube, out through the side of the glass tube at s, and doAvn to 
the water-bath, where its extremity joins the second platinoid wire at m, the small 
lump of copper just mentioned. The two platinoid wires, pc?, one coming from the 
centre of the cooling globe, and the other from the junction are carried forward 
to 0 , where they are soldered to the copper Avire electrodes of a reflecting galvano¬ 
meter to be spoken of immediately. It is necessary that no electromotive force, due 
to differences of temperature, should arise at these joinings. These joinings are, 
therefore, wrapped in tissue paper for insulation ; and then a large quantity of cotton 
Avool, 0 , is wrapped round the whole and along the Avires for some distance. The 
protection and equable distribution of temperature is so perfect that the cotton avooI 
bundle may be handled freely Avith wanned hands Avithout the galA'anometer being 
in any way disturbed. 
The galvanometer used in these experiments is a Thomson’s reflecting galvanometer. 
It has a very excellent coil, 0'9 ohm resistance, specially wound for me for use with 
thermo-junctions, by James White, GlasgOAv. 1 use a Steinheil plane parallel 
mirror, hung by a spider line instead of silk fibre, in a Thoaison’s dead-beat plug. 
The comfort and satisfaction of a spider line for purposes Avhere the galvanometer 
deflection is to be read, and not merely a zero noted, or Avhere a steady unchanging 
zero is desired, are almost indescribable.* 
Placed opposite the mirror is a Steinheil telescope and a scale. Every part of the 
galvanometer apparatus is screAved down on a Avooden bracket fixed to the stone 
Avail of my experimenting room, and the field magnets are fixed doAvn firmly Avith 
cement. I use a small Aveak “ zero magnet,” Avith its magnetic axis perpendicular to 
* .T. T. Bottomlet and A. Tanakadate. ‘ Roy. Soc. Proc.,’ \"ol. 4G, 1889, p. 28G. 
