GASES BY THE ACTION OF RONTGEN RAYS AND OTHER AGENTS. 
425 
apparatus a tube of glass wool was inserted, to prevent any of the small particles 
passing over with the oxygen. 
The apparatus was kept at a low pressure hy the pump, and a stream of oxygen 
made to pass through it hy heating the permanganate tube. The gas passed through 
the mercury in B on its way to the pump. Th.e connection to the pump could be 
closed at any time and the space above the mercury in B connected with the atmosphere. 
The mercury then rose in the long vertical tube, which dipped into it to a depth of 
about 1 centim.,and noAv served to indicede the pressure. The pressure could now be 
raised to any desired amount, less than that of the atmosphere, hy further heating of 
the permanganate. By closing the connection between B and the atmosphere and 
opening that leading to the pumj), the apparatus could he again exhausted and a 
stream of oxygen allowed to pass through it at a pressure of a few centimetres of 
mercury. 
A quartz lens, fixed above the quartz plate, served to bring the light fi’om an arc 
to a focus a little below the middle of the tube. The arc was formed between two 
horizontally placed carbons contained in a box with an aperture below somewhat smaller 
than the diameter of the tube. 
A current of oxygen was passed through the apparatus, while this was connected 
to the pump for 15 minutes, the apparatus was then left for one night and oxygen 
again allowed to stream through it for 30 minutes. After the pressure had been 
raised to 70 centims. the contents of the tube were exposed to the ultra-violet light 
of the arc. A fog appeared in a very few minutes. The apparatus was again left for 
a day and then pumped out, and a stream of oxygen allowed to pass for 10 minutes. 
Again, less than two minutes’ exposure to the ultra-violet rays was sufficient to produce 
a fog. A gain, after standing for three days, while repeatedly exposed to tlie ultra¬ 
violet rays till a fog a]q)eared, the oxygen Avas })um|)ed out and a vigorous stream 
passed for 30 minutes. The })ressure Avas then lArought up to 50 centims., and the 
gas exposed to the ultra-violet rays. A fog apjieared after an exposure of about one 
minute. 
The presence of nitrogen thus appears to he unnecessary for the production of cloud 
by the idtra-violet rays. There is no indication of any diminution in the density of 
the clouds or in the ease Avith Avhich they are produced as the gas becomes purer. 
The quantity of matter in the clouds Avliich develo}) under the action of ultra-violet 
light is very small: as is seen from the fact that even isolated patches of the fog 
remain suspended in the tube. Since tlie mass of each droj), even if its diameter be 
as great as one mean AA^ave-length of liglit, does not amount to 10“‘' gram, a very 
large number of drops may he present although the total Aveight of the fog is very 
small. 
The small quantity of matter in these clouds makes it A^ery difficult to exclude the 
possibility of their formation being due to the presence of traces of some vapour, 
which might become oxidised under the influence of the ultra-violet rays. That it is 
3 I 
VOL, CXCII.—A, 
