342 
DR. MEYER WILDERMAN OX CHEMICAL DYNAMICS 
latter is also connected with the hulb (18), (21) and (22) being mercury taps. The 
bulb (18) is connected with a manometer (E') and is filled first with carbon 
monoxide. The pressure of carbon monoxide in (18) should be greater than the 
pressure of chlorine in the quartz vessel (11). By opening the tap (21) the carbon 
monoxide is allowed to pass into the quartz vessel (II). No chlorine, as direct 
careful tests showed, passes under such circumstances into the bulb (18). 
I now pass to a more detailed description of the apparatus used. 
The Reaction Vessel (fig. II.), was a glass cylinder of 3’65 centims. diameter, 
3‘6 centims. long (capacity = 4R2G cub. centims.), with flanges 6 miflims. wide at 
each end (as in fig. II.). The flanges were very carefully ground so that the two 
perfectly smooth quartz plates, when placed on them, fitted perfectly, and an 
excellent vacuum could be obtained. The quartz plates, with parallel surfaces, 
45 sq. millims., and about 2 millims. thick, were cut out perpendicularly to the optical 
axis, and were optically pure. The front plate turned the plane of polarisation to 
the left, the back to the right. 
Since chlorine acts on any cement which could be used for the purpose of keeping 
a vacuum, the quartz plates were placed directly on the well-ground flanges of the 
glass cylinder, and melted Crookes’ cement, consisting of beeswax and resin 
(proportions, 5 parts beeswax to 8 parts resin) put on outside where the outer edges 
of the quartz plates met the glass flanges, care being taken that the cement was 
properly melted and free from air. The corners of the quartz plates projecting over 
the flanges and those parts of the glass flanges which remain uncovered by the 
quartz plates were then covered up with melted Crookes’ cement, the whole being 
heated and made smooth and firm by a very small flame. In this condition the 
vessel can preserve a very high vacuum for a very long time. The cement was then 
covered with a varnish of pitch in benzine to protect the cement from the action of 
water, and allowed to become thoroughly dry The outside of the cylinder was also 
covered with pitch so that light might only enter through the quartz plates. 
In the capillary tube (23), at the bottom of the glass cylinder, an iron-nickel couple 
was fixed to measure the temperature of the gas in the inside of the cylinder during 
the reaction. Two very fine iron-and-nickel wires (about 0'1 millim. diameter) were 
made considerably thinner at their ends -(0'05 to 0‘02 millim.) by repeated alternate 
immersions in nitric acid and in water. About 1 centim. of the two very fine ends 
were twisted together and a trace of melted soft solder put on the extreme ends. On 
bringing it near a flame, the solder melted and ran down the twisted wires and was 
quickly shaken off, forming only a very thin film of solder between the wires. The 
soldered ends were then washed and all but 2 millims. of the connected ends were 
removed. The two wires were then covered with shellac and a thin layer of pitch. 
The very thin double wire (perhaps 0‘2 millim.) was then passed through the very 
narrow capillary tube (23) projecting about I millim. into the cylinder, and fixed at 
the bottom in the capillary tube by a trace first of pitch, then of Crookes’ cement, 
