from an Arc and from Incandescent Metals. 251 



greater quantity, and the amount of negative was very little less 

 than that of positive. We give the numbers for one such experi- 

 ment ; the rate of leak of the terminal B (Fig. 1) is measured 

 when charged positively and negatively without the blast directed 

 on the arc and with it. The flow of gas past the terminal B is 

 the same whether the blast is used or not. 



Without the blast. 

 Positive charge 35 scale divisions in 30". 

 Negative charge 180 scale divisions in 30". 



With the blast. 



Positive charge 200 scale divisions in 30" (approx.) 

 Negative charge 200 scale divisions in 30" (approx.) 



With the blast therefore the amount of negative which comes 

 off in the gas was approximately the same as that of the positive, 

 while without the blast the amounts were in the ratio of 1 to 5. 



This experiment would seem to favour the explanation ad- 

 vanced above for the great difference in the amounts of positive 

 and negative carriers usually observed when C0 2 is used. 



(15) The velocity of the carriers in C0 2 varied with the 

 current through the arc in the same way as in air and oxygen ; as 

 the current is increased the velocity of the carrier rapidly dimin- 

 ishes. Thus in one series of experiments in which the current 

 through the arc was gradually increased, values of the velocity 

 were found diminishing from "01 cm. per second to "004 cm. per 

 second with a force of 1 volt per cm. No doubt a much greater 

 range of values could be obtained by varying the current through 

 the arc through a wider range. These determinations are all 

 made after the gas has come away from the arc and when it is at 

 the temperature of the room. 



(16) Comparing the velocity of the carrier when C0 2 is used 

 with that in air, the arc being as nearly as possible the same in 

 both cases, we always get a much smaller velocity in C0 2 than in 

 air. The ratio of the velocity in C0 2 to that in air was found in 

 different determinations with different arcs to vary from £ to T ^. 

 In C0 2 therefore the carrier finally collects around itself a greater 

 mass than it does in air. 



(17) In C0 2 as in air the velocity of the negative carrier is 

 greater than that of the positive ; the difference of velocities 

 varied considerably in a number of determinations made with 

 different arcs, but the mean value of the difference amounted to 

 about 40 per cent. The difference was therefore greater than in 



