A technique which led to the measurement of still longer wave lengths is the method of focal isolation employed by Rubens and R. W. Wood. The radiation from a suitable source is allowed to pass through a quartz lens, quartz being very transparent and also particularly refracting (refractive index 2.2) for the very long rays. The distance of the source from the lens is so arranged that while the shorter heat and the light waves actually diverge after leaving the lens the long heat rays come to a focus, which is iso lated by a hole in a screen. A stop at the centre of the lens pre vents the direct passage of practically unrefracted rays through the hole. To purify the radiation still further a second quartz lens with a central stop is used to refract the radiation passing through the hole. The wave-length is measured by an inter ferometer with quartz plates, maxima and minima of intensity being obtained. With a Welsbach gas mantle wave-lengths of 107µ or so were measured. Replacing the gas mantle by a quartz enclosed mercury vapour lamp Rubens and von Baeyer (19I I) found two bands of very long wave-length, one at about 218A and the other at 34211. They used an interferometer of Boltzmann type for the analysis of the radiation. Later (1921) Rubens used a grating made of wires a millimetre thick, with millimetre inter spaces, to disperse the radiation into a spectrum, and confirmed the presence of strong radiations at these wave-lengths, with measurable radiation extending to wave-lengths as long as 400/2. It is probable that the radiation of the mercury lamp is not a temperature radiation (see HEAT), but due to the motion of charged ions ; for instance Lindemann has suggested that positively and negatively ionised atoms, separated by about the same dis tance as the atoms in liquid mercury, may form doublets which rotate like double stars, and has shown that this would lead to wave-lengths of the order required.
As regards radiation from a black body of the hollow-space (Hohlraum) type used by Lummer and Kurlbaum (see BLACK BODY), Rubens and Michel made detailed measurements on a radiation as long as 52p, from such a source. It is therefore clear that the gap between Hertzian waves produced by discharge be tween microscopic particles on the one hand, and infra-red rays produced from hot solids and vapours on the other hand has been closed with an ample overlap, the Hertzian waves extending down to Ioc.4, and the infra-red rays reaching up to 400/1.
The first great step in the modern extension of the ultra-violet was taken by Schumann, who, by using fluorite lenses and prisms and special plates, in which the emulsion was deliberately made very poor in gelatine, since gelatine has a marked absorption for the far ultra-violet, pushed the recorded limit to 1,220 A. He showed (1893) that radiations between i,800 and 1,200 A. are strongly absorbed by air, and hence that it was necessary to work with the spectrograph in vacuo. (See SPECTROSCOPY.) At about 1,200 A.U. fluorite becomes opaque to the radiation, so that for his work on the still shorter waves Lyman, whose researches extend from 1906 to the present day, used a concave grating without lenses or mirrors, the grating itself forming images of the slit on the photographic plate. The spectrograph was necessarily enclosed in a vessel which could be evacuated. Lyman has reached 5oo A. During the years 1923-1928 Millikan and his collaborators made further very important advances into the extreme ultra violet. The source used is the so-called "hot spark" between elec trodes of the metal whose spectrum is being investigated : the spark gap is very short and situated in a highly evacuated space, of pressure .0000i mm. of mercury or less, and the discharge from a powerful battery of condensers charged by an induction coil is passed intermittently. Specially ruled gratings are used, and
the pressure in the whole apparatus maintained at the lowest pos sible level. The limits reached with the line spectra of certain elements are as follows: carbon 360.5 A.U., zinc 317.3 A.U., iron 271.6 A.U., silver 26o A.U., nickel 202 A.U. and aluminium 136.6 A.U. This 136.6 A.U. represents the extreme short wave limit measured by the ordinary methods of optical, as distinct from X-ray, spectroscopy.
Turning to the advance from the X-ray side, the longest waves measured by the methods of crystal diffraction originated by the Braggs are about 13 A.U., e.g., La for copper is of wave-length 13.3 A.U., while one of the N series lines of thorium, which seems to represent the limit, is of wave-length 13.8 A.U. The measure ment of rays in the gap between 13.8 A.U. and 136.6 A.U. presents peculiar difficulties, because all the wave-lengths in ques tion are particularly easily absorbed, even by gases at low pressure, and because the diffraction methods break down in this region. As an example of the absorption it may be mentioned that Holweck finds that a sheet of celluloid .00027 mm. thick—and celluloid has a particularly low absorption coefficient in this region—transmits only 3 per cent of the radiation in the neighbourhood of 310 A.U., where the absorption is at a maximum. It is consequently neces sary to maintain a very good vacuum in the apparatus, and, if any solid diaphragm is used, to make it exceedingly thin. The rays are usually detected either by their photoelectric effect (see PHOTO ELECTRICITY) or by the ionisation which they produce in a gas at low pressure.
McLennan, Horton, Hughes and others have employed various forms of apparatus in which the surface of a solid is bombarded by electrons accelerated by a measured difference of potential, the radiation so excited being detected and measured by its photoelectric effect on a metal plate. The highest possible vacuum is maintained in the apparatus, so that the incident electrons have the full energy corresponding to the voltage drop. A curve con necting this voltage with the photoelectric current released by the radiation is obtained, and sharp changes in direction of this curve are taken to indicate the appearance of a new radiation, whose wave-length can be calculated from corresponding voltage. Such a voltage is a resonance potential (q.v.). The plate from which the photoelectric current is measured has to be very carefully shielded from any diffusion of ions by arrangements of charged gauzes which vary in disposition in the apparatus of dif ferent workers. The method has also been applied to gases, which are bombarded by electrons of known energy. (See RESONANCE POTENTIAL.) Holweck, on the other hand, has employed a differ ent method, separating the part of the vessel where the short waves are produced from the part where their effect is measured by a solid diaphragm, which stops all disturbing ions. The celluloid films which he uses for this purpose have a thickness of cm.
or less. The intensity of the radiation is measured by the ionisa tion produced in a gas, the pressure of which varies for different experiments, but is generally of the order of a few millimetres of mercury, as contrasted with pressures of the order of .00005 mm.
of mercury in the generating part of the apparatus. A sudden change in the ionisation corresponds to the attainment of a reso nance potential, from which the wave-length is calculated as in the other type of experiment. The apparatus can also be used to measure the absorption of these extremely soft (i.e., easily absorbed) radiations.
The result of these measurements of resonance potential is to reveal the existence of characteristic radiations from given ele ments, of the nature of an extension of the X-ray line spectra (see