632 REPORT— 1895. 



a measured quantity of the gas ^ from the gas-holder through an indiaruhber tube 

 to a receiver of known efficiency and of known capacity in connection with the 

 electrometer. We have not yet measured how much electricity was lost in the 

 passage through the indianibber tube. It was not probably nothing ; and the 

 electric density of the gas before leaving the gas-holder was no doubt greater, 

 though perhaps not much greater, than what it had when it reached the electric 

 receiver. 



§ 7. The efficiency of the electric receivers used was approximately determined 

 by putting two of them in series, wdth a paraffine tunnel between them, and mea- 

 suring by means of two quadrant electrometers the quantity of electricity which 

 each took from a measured quantity of air drawn through them. By performing 

 this experiment several times, with the order of the two receivers alternately 

 reversed, we had data for calculating the proportion of the electricity taken by 

 each receiver from the air entering it, on the assumption that the proportion taken 

 by each receiver was the same in each case. This assumption was approximately 

 justified by the results. 



§ 8. Thus we found for the efficiencies of two different receivers respectively 

 077 and 0'31 with air electrified positively or negatively by needle points ; and 

 082 and 42 with carbonic acid gas electrified negatively by being drawn from an 

 iron cylinder placed on its side. Each of these receivers consisted of block tin 

 pipe 4 cm. long and 1 cm. diameter, with five plugs of cotton wool kept in position 

 by six discs of fine wire gauze. The great diti'erence in their efficiency was, no 

 doubt, due to the quantities of cotton wool being dill'ereut, ur differently compressed! 

 in the two. 



§ 9. AVe have commenced, and we hope to continue, an investigation of the 

 efficiency of electric receivers of various kinds, such as block tin, brass, and 

 platinum tubes from 2 to 4 cm. long and from 1 mm. to 1 cm. internal diameter, 

 all of smooth bore and without any cotton wool or Avire gauze filters in them ; alsa 

 a polished metal solid insulated withhi a paratfine tunnel. This investigation, 

 made with various quantities of air drawn through per second have already given 

 us some interesting and surprising results, which we hope to describe after we 

 have learned more by further experimenting. 



§ 10, In addition to our experiments on electric filters we have made many 

 other experiments to find other means for tlie diselectrification of air. It might 

 be supposed that drawing air in bubbles through water should be very effective for 

 this purpose, but we find that this is far from being the case. We had previously 

 found that non-electrified air drawn in bubbles through pure water becomes nega- 

 tively electi'ified, and through salt water positively. We now find that positively 

 electrified air drawn through pure water, and negatively electrified air through 

 salt water, has its electrification diminished but not annulled if the primitive elec- 

 trification is sufficiently strong. Negatively electrified air drawn in bubbles through 

 pure water, and positively electrified air drawn through salt water, has its electri- 

 fication augmented. 



§ 11. To test the effects of heat we drew air through combustion tubes of 

 German glass about 180 cm. long and 24 or H cm. bore, the heat being applied 

 externally to about 120 cm. of the length. AVe found that wlien the temperature 

 was raised to nearly a dull red heat, air, whether positively or negatively electrified, 

 lost little or nothing of its electrification by being drawn through the tube. When 

 the temperature was raised to a dull red heat, and to a bright red, high enough to 

 soften the glass, losses up to as much as four-fifths of the whole electrification were 

 sometimes observed, but never complete diselectrification. The results, however, 

 were very irregular. Non-electrified air never became sensibly electrified by being 

 drawn through the hot glass tubes in our experiments ; but it gained strong posi- 



' The gas-holder was 38 cm. high and 81 cm. in circumference. Ten strokes 

 of the pump raised the water inside tu a height of 8"! cm., so that the volume of 

 air drawn througli the receivers in the experiments was 428 cubic cm. per stroke 

 of the pump. This agrees with the measured efEective volume of the two cylinders. 

 of the pump. 



