JANUARY 27, 1923] 

X-ray wave-lengths so far as possible in the Soa, wave | the ‘one 
direction by crystal methods, but it is evident that 
there must be a limit, and it is possible that this limit 
has almost been attained, for in spite of the great 
improvement 
Ye00 
3968, 1880 1F20 B00 
= : n ea 
A INFRA Reb Visi@le] Quarrz |Fevo} Z 
RrEl es} | 
NATURE 
in technique and the extraordinary | corresponding 

119 
line, the numbers given at the top being 
corresponding wave-lengths in Angstrém units. It 
will be seen that this representation is similar to that 
of the keyboard of a piano, equal horizontal spacings 
everywhere to an equal number of 
a oe SSS ee 
Vis) BLE 
VisiBUE 
8 
activity in this line of work since Moseley’s measure- 
ments in 1913, the longest wave-length I have been 
able to find recorded as measured is the zinc L,, line 
given by Friman as 12-346 A. This represents but 
half an octave out of the seven octaves between the 
limits left by Lyman and by Moseley. 
The failure of crystal methods is due to two causes. 
The distance between the centres of the atoms in solids 
is of the order of an Angstrém unit, so that at 12 A 
the waves are already much longer than the distance 
between the reflecting planes which form the grating 
elements. (For the crystals rock-salt and calcite, with 
which most of the accurate measurements have been 
carried out, these distances are 2°184x1078 and 
3°028 x 10-® cm. respectively.) The other difficulty 
arises from the intense absorbability of these soft 
X-rays by practically everything, a phe nomenon that 
we have already witnessed in the radiation the 
other side of the gap. Sir William Bragg has recently 
been investigating some organic crystals which have 
grating spaces very much farther apart than rock-salt 
and calcite, and it may be that in employing such 
crystals in an evacuated system we have a way of 
making considerable advances into the gap from the 
high- frequency end by the X-ray crystal diffraction 
methods. It would seem that in Moseley’s original 
apparatus we have an arrangement which could be 
rather easily developed for this purpose. Another 
advantage of these crystals is the possibility that they 
may not “absorb the rays so very intensely, as the only 
known substances which have appreciable transparency 
in this region are organic compounds or mixtures of 
them, such as celluloid. 
Returning to the position about 1973, ie is con- 
veniently exhibited by diagram A of Fig. , in which 
the various spectral limits are marked tant an even 
scale aang to the logarithms of the correspond- 
ing frequencies. These are shown by the numbers on 
io 9778, vor. 111] 
on 

octaves. The great width of the gap between the 
X-ray and ultra-violet limits is very apparent. 
A very considerable advance into this gap was made 
by Dr. Bazzoni and myself in 1917 using a method 
which was novel in spectroscopy. Our experiments 
were directed towards the measurement of the short 
wave limit of the arc spectra of various gases, and 

ction of apparatus used for the meas 
igth limuat 

f arc spectra, drawn to s 
more particularly of helium, which are generated when 
such bombarded by considerable electron 
currents under moderate voltages (Fig. 2). The radia 
tion from the generated under impact of the 
electrons passing from the incandescent tungsten 
cathode F to the cold anode Aj, falls on the metal strip 
T, after through the gap between the metal 
plates P, across which an electric field is maintained 
of sufficient strength to remove any ions present in the 
radiation stream, This radiation liberates electrons 
gases are 
gas 
passing 
