208 
MR. A. Ll, hughes ON THE 
film of gas could be formed, even in vacuo, in a few seconds after the exposure 
of a new surface. To meet this objection a more drastic treatment was devised. 
Velocity measurements were made while the mercury was in motion. The tap T was 
adjusted so that the mei'cury overflowed at C at the rate of about 3 c.c. per second. 
On considering this, together with the dimensions of the dish, it is clear that no part 
of the surface exists for more than a small fraction of a second. Owing to the fact 
that the mercury in D was soon exhausted, accurate measurements could not be made. 
However, it can be said that the maximum emission velocity from the continuously 
forming mercury surface was 2T5±'05 volts. This is practically the same value as 
that obtained for distilled Zn, Cd, and Hg. It seems extremely unlikely that any 
retarding film could be foi'ined on the Hg surface in this experiment, hence it is 
concluded that 2T5 volts represents the actual emission velocity from Hg. The 
agreement with the distillation experiments supports the view that metals prepared 
by distillation in vacuo are also free from surface films. 
In the course of this exjDeriment evidence was obtained which indicates that photo- 
electrons are emitted nearly equally in all directions. The photo-electric current from 
C to A was measured in the absence of an electric field between C and B—(l) with, (2) 
without, a strong magnetic field along the axis of the cylinder. In the first case, when 
the electrons were compelled to follow the lines of magnetic force, the current was 33 
times larger than when the electrons were not constrained to follow any path. The 
ratio of the solid angle subtended by the gauze B at the mercury surface to 27r, was 
about 1 to 38. This indicates that the ratio of the electrons emitted inside the cone 
to those emitted outside is of the order of the respective solid angles. 
From the experiments described and discussed in this section, it was concluded that 
the most satisfactory surfaces to use in the investigation on the velocities of the photo¬ 
electrons were those prepared by distillation in vacuo. 
4. The Source of Light .—The mercury arc in quartz glass was used as the source of 
light in this research. In the early stages of the work a very intense spark between 
A1 terminals, produced by the primary transformer of a Tesla apparatus, was used in 
order to get very short wave-lengths of appreciable intensity. The results obtained 
with this intense spark were most irregular, frequently emission velocities as high as 
10 or 12 volts were obtained. Direct experiment showed that this was not an effect 
due to electric waves. The spark was finally abandoned in favour of the mercury arc 
which gave very regular results, although the energy available in the short wave¬ 
lengths was far less. 
To produce monochromatic light, a Hilger quartz sjDectrograph was adapted for the 
purjDose. A sliding slit was arranged to pass over the focal plane of the camera side 
of the instrument. The slit having been set so as to transmit any desired wave¬ 
length, tlie mercury arc was placed close to it and monochromatic light emerged from 
the collimator. The aperture of the slit was 14 mm. by 2 mm., the collimator lens was 
2 cm. in diameter, and of 20 cm. focal length. Hence the amount of light which 
