452 SECTIONAL TRANSACTIONS.—At. 
The necessity for avoiding ‘ cracking ’ restricts the maximum temperature 
and the pressure of the vapour stream, and the fact that the oil wets the 
pump surfaces prevents the use of small clearances between cowl and 
condensing tube. It is therefore not possible to use such pumps against 
a high fore vacuum pressure. Consideration of the properties of a pump 
as a vacuum fractionating column shows that the extent to which the final 
vacuum is influenced by impurities of small amount (1 in 10°) may depend 
markedly on whether the condensing surface is water-cooled or not, even 
though a water-cooled trap is placed between the pump and the fine system. 
Similar considerations show that it is not advisable to cool the fore vacuum 
pipe. It is necessary to constrict the pump mouth above the cowl, to con- 
dense the ‘ reverse jet’? due to multiple collisions : this theoretical point 
appears to have escaped notice so long as mercury was the only working 
fluid used. 5 
Contamination of the working fluid by mercury has been found to be an 
insidious cause of trouble, so that it is desirable to modify one’s technique of 
performing various experiments—e.g. speed measurements—-so as to avoid 
the use of mercury. It is in any case undesirable to use McLeod gauges 
in oil condensation pump systems, as the readings are no criterion that the 
pump is working properly. It is necessary to bake out an ionisation gauge 
before an oil condensation pump will produce a really good vacuum in it ; 
it is difficult to say how much this is a question of outgassing the gauge, 
and how much a question of indirect ‘ conditioning ’ of the pump. 
Mr. B. L. Goopier and Mr. A. BEETLESTONE.—The production of 
high voltages. 
The paper deals with the production of small powers at high voltages for 
general laboratory purposes, and particularly for use in connection with 
high vacuum equipment. 
Such high voltages may be either unidirectional or alternating. Uni- 
directional voltages may be continuous pulsating or impulsive ; alternating 
voltages may be damped or undamped and of high or low frequency. ‘The 
paper outlines the characteristics and methods of producing these various 
types of high voltage, and gives details of equipment available. 
Dr. T. E. ALLIBonE.—High voltage vacuum tubes. 
This paper deals with the construction of vacuum apparatus suitable for 
operation at high voltages, such as electron and positive ion discharge tubes, 
oscillographs, thermionic rectifiers, three electrode valves, and X-ray tubes. 
The apparatus described is primarily intended to be continuously 
evacuated, and demountability is one of its special features, so facilitating 
experimental work and the replacement of defective parts. Even for high 
voltage experiments the oil diffusion pumps are operated without the use 
of liquid air traps, thus allowing big pumping speeds to be attained. 
Examples of high voltage vacuum tube pumping systems are given. 
Dr. J. D. Cockcrort.—A high voltage D.C. generator. 
In order to produce high speed atomic particles for work on nuclear 
transformations, it has been necessary to develop sources of steady potential 
of seven or eight hundred kilovolts, capable of producing currents of several 
milliamperes, and to develop vacuum tubes to withstand these potentials. 
Transformers capable of providing these potentials are bulky and expensive, 
and difficulties arise in applying such transformers directly to a chain: of 
rectifiers. A voltage multiplier has therefore been devised which will 
