SCIENCE AND INDUSTRY. 
Ordinary radiography depends entirely upon the difference in absorbing 
powers of different materials. It was found by Réntgen himself and con- 
firmed by later observers that this absorption increased with the density of 
the material. More precisely, the absorption by any elementary substance 
is proportional to the number of atoms penetrated and to the cube of the © 
number which gives the place of that element in the periodic system. ‘Thus 
an atom of calcium, the twentieth element, absorbs (20/8)? = sixteen times 
as much as an atom of oxygen, place 8. It is for this reason, e.g., that lead 
is so much employed for protective screening in X-ray work. This absorp- 
tion is due to two main causes: firstly, a part of the radiation is used up in 
exciting the characteristic or “fluorescent” radiation of the absorbing 
element, and secondly a certain fraction is scattered or diffused in a manner 
similar to that in which the sun’s radiation is scattered in passing through 
the earth’s atmosphere. Now fluorescent radiation can be excited only by 
radiation of wave-length shorter than its own. Hence a curious irregularity 
appears in the relation between absorption and wave-length of radiation. 
If we start with a wave-length considerably below that characteristic of the 
absorber, and gradually proceed to shorter and shorter wave-lengths, the 
penetration as we would naturally expect will steadily and very rapidly 
increase, the fraction absorbed as rapidly decrease. But so soon as the 
wave-length of the radiation is diminished to that at which the characteristic 
_ tadiation of the absorbing substance is produced, a large increase in absorp- 
tion occurs. This fact is of importance to the radiographer, for photographic 
action is dependent upon absorption of the radiation by the sensitised 
emulsion, and it is seen that we do not necessarily secure greater effect by 
using a softer radiation. Fig. 6 illustrates the increase in photographic 
action occurring when the wave-length becomes short enough to excite the 
characteristic radiation of silver. 
A lien ~BrAg 97 %8 
FIG. 6.—Spectrum of X-rays, showing sudden increase of absorption at a point 
where the fluorescent radiations of silver and of bromine in the photographic 
a emulsion are excited. ng 
This fact of greatly increased absorption at a certain critical wave-length 
can be made use of to secure a radiation in which one particular wave-length 
greatly predominates over all others. If, for example, the radiation from a 
tungsten tube operated at eighty to ninety thousand volts or higher, and 
producing strongly the characteristic K-radiation of tungsten of wave-length 
.212A., be passed through a filter of ytterbium or any compound of this 
element of properly calculated thickness, the radiation thus “ filtered” is 
one in which the characteristic radiation of tungsten amounts to a very 
° / 
532° 
