146 Report §.A.A. Advancement of Science. 



base of the iron cylinder and on the ends of the pipes (A) and (B) 

 next the cylinder. The flame was placed underneath the base of the 

 cylinder. 



A quadrant electrometer, surrounded by wire gauze, was used 

 in both arrangements. The lilter (F,.) was inside the gauze and in 

 permanent metallic connection with one pair of quadrants. This 

 pair could be joined to case or insulated at will. The metal tubes 

 and cylinders, the hlter (FJ, the wire gauze and the other pair of 

 quadrants were metallically connected with the case of the instru- 

 ment. The axial wire in the first arrangement, or the stout brass 

 wire in the second, could be connected with the case of the instru- 

 ment or with the source of a constant E.M.F. just as desired. 

 It may be stated once for all, that, before each experiment, the electro- 

 meter, with connected hlter (F,,), was tested for insulation, and 

 no results were taken when the insulation was faulty. Tests were 

 always made to ascertain that the air passed from the room through 

 the filter (F.,), through the metallic vessels, through the filter (F^), 

 and thence to the pump. In no case was any trace of electrification 

 obtained on drawing air from the room in this way when the flame 

 was not lit. Great care was also taken to prevent products of com- 

 bustion from entering the metallic vessels. 



§3. Preliminary Experiments with an iron tube, a copper tube, 

 and with phosphorus in a zinc tube, the arrangement being as in 

 Fig. IV. 



In the case of the iron and the copper tubes, six burners 

 in the furnace were used, and the flame was applied directly. The 

 tubes were heated to a cherry red. In the case of the iron tube, it 

 was found that, with everything to case, the air drawn through was 

 positively electrified when tested by the filter (F,.). The amount of 

 the electrification increased with the electromotive force. The 

 nature of it was positive, quite irrespective of the nature of the charge 

 given to the insulated axial wire. The electrification was not 

 removed by drawing the air through (i.) a U-tube filled with pumice- 

 stone soaked in strong sulphuric acid, (ii.) a wash-bottle containing 

 water, (iii.) first water and then a solution of potassium iodide in 

 water, or (iv.) an alkaline solution of pyrogallic acid and a U-tube 

 containing pumice-stone soaked in strong sulphuric acid. It was 

 removed so far as the filter (Fc) was able to test by drawing the 

 air through water, potassium iodide solution, pyrogallic acid and the 

 U-tube, before reaching the filter. 



It is to be remembered, however, that the effect of passing the 

 electrified air through solutions of this nature is twofold, (a) certain 

 constituents may be taken from the air, or (b) the bubbling may cause 

 an electrification of the air sufficient to give an indication on the 

 filter. This latter effect was allowed for by drawing non-electrified 

 air through the various solutions, and then testing by the filter. The 

 result was a slight negative electrification, causing a deviation of a 

 fraction of a scale division on the electrometer. 



The following table shows the effects produced. 



The figures are the mean of a nimiber of experiments. 



