DR. HAiiOLD A. WILSON ON THE 
•24G 
experiments the electric intensity near the hne wire approaches this value even when 
the P.D. is only 100 volts or less. 
McClelland states in this paper that the negative leak from the wire is nearly 
independent of the air pressure from oC millim. to niillira., and he suggests that 
the negative leak at low pressures is caused by the escape of negative ions from the 
metal. 
Mr. O. W. lliCHARDSOX (‘ Proc. Camh. Phil. Soc \ol. 11, Part IV., 1902, p. 28 g) 
has published the results of a series of measurements of the negative leak from hot 
platinum in gas at low pressures. He finds that the variation of the leak with the 
temperature can he represented by a formula of the type x = Thus formula 
PiCHAE-DSON deduces from the assumption that the platinum contains negative 
corpuscles which can only get through the surface layers when their velocity exceeds 
a definite value. The value he obtains for Q corresponds to a fall of the ionic charge 
throup’h 4T volts, which does not differ much from the value 2’6 volts found hv the 
writer in air at atmospheric pressure. According to PvTCHAEDSOn’s theory, Q is the 
energy required to drive a gramme molecular weight of the corpuscles out of the 
metal through the surface layers. It is shown below that the formula x = 
can be deduced without making any further assumption than that ions are produced 
at the surface of the metal, so that tlie agreement of the experimental results 
obtained with this formula cannot he regarded as evidence in favour of any particular 
theory with regard to the way in which the ions are produced. Richaedson also 
shows that on his theory it is possible to deduce from the constant A in the above 
equation the number (n.) of corpuscles in a cubic centimetre of the platinum. He 
obtains n = 1'3 X lO'b which is of nearly the same order of magnitude as the values 
of }i deduced from experiments'^ on the variation of the electrical resistance oi 
})latimim in a magnetic field. Richaedson has recently read a paperf before the 
Royal Society on this subject, in which he has examined the negative leak from some 
other metals besides platinum in gas at low pressures. The results with platinum 
descrilied in that paper are the same as those just referred to. In all these 
experiments the positive leak was very small compared with the negative leak. 
W. Wien (‘ Annalen der Physik,’ vol. 8, 1902, p. 244) has described some 
experiments on the leak from hot platinum wires in high vacua. He finds that even 
in a very good vacuum a hot wire can still discharge electricity to a distant small 
electrode if it is charged to a rather high potential. In the best vacuum the P.D. 
necessary was 800 volts for negative electricity and 3000 volts for positive. 
According to 0. D. Child (‘ Phys. Review,’ Nos. 4 and 5, 1902) the negative leak 
from a hot })latinum wire, using a large P.D., increases very rapidly with diminishing 
pressure from one atmosphere down to a small fraction of a millimetre of mercury. 
It will be shown below that this large increase is due to ionization by collisions with 
* Patterson, ‘Phil. Mag.,’June, 1902. 
t >See ‘Phil. Trans.,’ A, vol. 201, 1903. 
