July 15, 1875] 



NATURE 



217 



CHARCOAL VACUA* 

 {From a Correspondent^ 



PROF. DEWAR began his discourse by describing 

 the different processes which have been adopted for 

 obtaining very perfect vacua, and referred to a paper 

 regarding this matter, read by Prof. Tait and himself 

 before the Society last year. 



By the ordinary air-pump the exhaustion can only be 

 obtained to \ of an inch, i.e. .^j^ of the ordinary pressure. 



Regnault, in some of his experiments, after exhausting 

 with the air-pump, boiled water, and when the water had 

 evaporated, sealed the vessel, and then broke a flask 

 inside containing sulphuric acid, and so the water vapour 

 was absorbed. 



Dr. Andrews' way is a revival of one due to Davy, viz. to 

 fill and exhaust twice with carbonic acid after the pump 

 exhaustion, and then by caustic potash to fix the CO2 

 which is left. 



Professors Tait and Dewar's method is to take advan- 

 tage of the power charcoal has of condensing gases ; 

 while the exhaustion, by means of a mercury pump, 

 is going on, the charcoal is kept heated ; when the 

 exhaustion has been carried as far as possible, the vessel 

 is sealed, and as the charcoal cools, it condenses the very 

 small residue of gas there may be present, and this can 

 again be temporarily driven out by heating the charcoal. 

 The test they have employed to gauge the perfection of 

 their vacuum has been to sec if it will allow an electric 

 spark to pass. It is well known that at the ordinary 

 atmospheric density it requires considerable tension for a 

 spark to pass through air, and as the density diminishes, 

 the spark passes more easily ; but when a certain point is 

 reached the difficulty again increases, and in a very per- 

 fect vacuum no spark passes at all. Two wires, \ inch 

 apart, in one of Tait and Dewar's exhausted tubes would 

 not allow a spark to pass, although a powerful coil was 

 employed. 



Prof. Dewar went on to say that the effect of light and 

 heat had been tried by many experimenters, on magnets 

 and delicately suspended bodies, and in the Edinburgh 

 New Philosophical yournal for 1828 there is an interesting 

 account of some experiments performed by Mark Watt on 

 the same subject, with apparatus little differing in appear- 

 ance from that now used by Mr. Crookes. 



Recently Mr. Crookes has found some curious results 

 which he seems to think are inexplicable. He found that 

 the action of a beam of light on a delicately suspended glass 

 fibre with a disc at each end was repulsion of the disc when 

 the exhaustion was perfect, but attraction when at ordi- 

 nary pressures. The discs were light bodies of pith or cork. 

 One side of each was covered with lampblack, the other 

 was white. The first thing to be noticed is that the 

 blackened face is affected much sooner than the white face. 

 Since there was attraction at one density and repulsion 

 at another, it follows that at some intermediate density 

 there is no action at all, and this neutral point depends 

 among other things on the conductivity of the body and 

 the nature of the residual gas. 



It will be seen that for delicate action one essential is 

 that the glass of the vessel be thin. The sensibility is 

 also found to increase with the perfection of the vacuum. 



The first fact ascertained is that the action follows the 

 law of the inverse square of the distance, that which all 

 radiation obeys. Thus, when the light was 3I inches from 

 the beam, the reading was no, zero 22, deflection 88; 

 at 7|, reading 48, deflection 22, or only about \ ; and 

 when at 1 1^, reading 33 ; and(as zero changed, reading 33, 

 deflection 9, or only about \. 



The next experiment was this. Professor Dewar in- 

 terposed between the candle and the beam a substance 

 opaque to heat rays. The candle was placed so as to 



* By Professors Tait and Dewar. Paper read by Prof. Dewar before the 

 R.S. of Edinburgh on Monday, July 12. 



give a large deflection, and then a vessel of ordinary 

 glass was interposed, and the deflection decreased, and 

 on filling the vessel with water, which is almost opaque to 

 heat rays, there was no perceptible deflection left. This 

 shows that when the heat rays are absorbed or prevented 

 from reaching the disc, hardly any action takes place. A 

 layer of water \ of an inch thick diminishes the amount 

 of deflection to \ part of the original. 



Next a smoked piece of rock-salt was interposed, or 

 a vessel filled with a substance transparent to heat 

 but opaque to light, viz., a solution of iodine in bisulphide 

 of carbon. The deflection was as before, large ; on the 

 empty screen being interposed a diminution followed, due 

 to the non-transparency of the glass screen for heat. 

 But when by means of the iodine solution the light rays 

 were cut off there was hardly any further diminution in 

 the deflection. This shows that the light rays may be 

 taken away without any considerable diminution of the 

 action. 



Prof. Dewar then proceeded to show that the heating 

 of the disc was the efficient cause of the action. 



Two equal discs, one of rock-salt, the other of glass, 

 were attached to; the glass fibre. The rock-salt was 

 inactive when the beam was thrown on it ; the glass disc 

 was active. The reason is evidently that the rock-salt 

 is not heated, being transparent to heat, whereas the 

 glass is opaque, absorbs the heat and is heated. Unless 

 the shell of the receiver be thin, however, the selective 

 action is very small, as the glass envelope absorbs much 

 of the heat. 



The back of the rock-salt disc was then coated with 

 lampblack, and the beam sent through to the blackened 

 side. Yet there would be attraction. The heat and light 

 passes through the rock-salt and is absorbed by the lamp- 

 black at the surface of contact. The lampblack is heated 

 up in consequence, but it is so bad a conductor that before 

 this heat can be conducted through the thin coating of 

 lampblack it is conducted through the rock-salt, heats it 

 up, and the action is repulsion. If the lampblack were 

 not so bad a conductor, all the lampblack would be first 

 heated up and there would be repulsion at the other side, 

 or apparent attraction. The subsequent action is due to 

 the giving out heat unequally on the two sides. 



The next modification was to substitute for the rock- 

 salt clear sulphur and ordinary sulphur on the other. 

 The peculiarity of clear sulphur is that when acted on by 

 light it resumes the appearance of ordinary sulphur, with 

 a disengagement of heat. A beam was thrown on this, 

 and the effect was, as expected, attraction, the back being 

 heated, and repulsion, there being attraction on the other 

 side. The success of this experiment depends on the 

 way in which the sulphur is transforming. 



This suggested to the learned Professors an instrument 

 for detecting the presence of very high violet rays. Have 

 the transparent discs coated with white phosphorus, 

 which is opaque to the ultra-violet rays. There would 

 ensue a chemical action with disengagement of heat, and 

 the result would be a motion of the discs. To show the 

 sensitiveness of the apparatus, it may be stated that an 

 ordinary lucifer-match held at a distance of 4 feet pro- 

 duced instant action, which was observed by means of a 

 telescope. When ether was brought near there was 

 attraction. The disc followed the ether about because 

 there was radiation of heat from the disc. The action is 

 clearly due to the infinitesimal heating of the discs. 

 Reynolds suggested the action was due to the evaporation 

 of some fluid on the surface of the discs. The recoil of 

 the evaporating particles leaving the disc sent it back. 



When the action takes place in ordinary pressures it is 

 probably due to convection currents. The air in front of 

 the disc is heated and ascends, there is avacuum,and hence 

 the disc advances. To understand the action that takes 

 place when the exhaustion is more perfect, we must con- 

 sider how much gas there is in the vessel. The capacity 



