194 F. Leininger: Relation of Electric Charges transported 
for reflexion yields impossibly large values, so that it appears 
to me unlikely that so great a proportion should undergo 
reflexion. 
It is also possible that a considerable proportion of the 
reflected particles again reaches the collector by conduction. 
TABLE VIII. 
Canal rays in hydrogen. Second kind of net-electrodes. 
Value of 
Value of 100x canal rays 
Potential in volts. 100 canal eye) current ” 
current corrected for 
absorption by net. 
300—500 12:2—13°9 

/ 
| 
} 
| 23-35 —25-26 
1000 | 16°8 | 32:4 
1200 | 185 | 35:7 
1400 | 19:2 / 37 
1600 | 20-6 | 39-7 
1800 | 22291 | 42-4426 
2000 23°7 45°7 
2400 | 25 48:25 
| 3000 | 26'8 | 51-7 
| 3600 | 28-4 54:8 
4000 30°3 58:5 
4800 | 32-2 62:1 
5400 33/1 63-9 
6000 | 34:3 66-2 

| 7000 35°6 68°7 
‘ 
, | 
Beyond 7200-8000 volts the discharge became irregular 
and disruptive, the influence-machine being obviously no 
longer capable of producing a steady discharge. For this 
reason the last values are mostly unreliable. 
On comparing the values obtained, we notice in the first 
place a somewhat striking absorption of the rays in oxygen, 
especially of the canal rays—an effect easily accounted for 
by the constitution of the particles in the latter case. It is 
further remarkable that at the higher pressures the percentage 
absorption of canal rays should be much higher than that of 
cathode rays, considering that the latter have to traverse a 
greater distance, and hence at the higher pressures their 
absorption becomes more important. 
From the above experiments with different gases filling the 
tube it follows that the absorption of the rays by the gas 
plays an important part. I therefore tried to devise some 
