526 
PHYSICS: DUANE AND PATTERSON Proc. N. A. S. 
The question now arisen as to whether or not an L critical ionization 
frequency equals an L critical absorption frequency. A critical ioniza- 
tion frequency represents the transfer of an electron completely outside 
of an atom. In the case of the K series of iodine the two frequencies have 
been found to be equal to each other within the limits of error of the 
measurements.^ Experiments are in progress to test this point for the 
L series. 
Relative Intensities of Lines. — The ionization spectrometer provides us 
with an excellent method of estimating the relative intensities of lines 
in X-ray spectra. Accurate estimates of the relative intensities, how- 
ever, cannot be obtained unless the lines lie fairly close together. If a 
considerable interval separates them, corrections must be applied for the 
changes with varying wave-length in the amounts of energy absorbed by 
the substances through which the rays pass, etc. These corrections can- 
not be accurately calculated at present. Special precautions must be 
taken to make sure that no critical absorption nor critical ionization 
wave-length lies between the wave-length of the lines to be compared 
with each other. These limits to the accuracy of relative intensity meas- 
urements present themselves in photographic spectrometry in addition 
to those due to the difficulty of estimating the blackening of the photo- 
graphic plates. 
In the case of a line that is not perceptibly broadened we take the 
height of the peak corresponding to it on the ionization current crystal 
table angle graph to represent its intensity. This height must be meas- 
ured from the level of the curve corresponding to the general radiation 
on the two sides of the peak, and not from the axis of zero ionization 
current. 
Two peaks appear on the curve in figure 1, corresponding to the spectral 
lines ^2 and The ratio of the heights of these two peaks is 116. This 
number, however, does not represent accurately the relative intensity of 
^2 to iSs, for the critical absorption of Lai lies between the two emission 
lines. The target absorbs a greater fraction of the X-ray energy in the 
neighborhood of ^5 than it does in the neighborhood of ^2- In this case, 
therefore, the ratio of the heights of the peaks gives us only an upper 
limit for the ratio of the intensities. ^2 cannot be more than 116 times 
as intense as 
The curve in figure 2 contains four peaks corresponding to the emis- 
sion lines 7i, 72, 73 and 74. The relative heights of these peaks are repre- 
sented by the numbers 100, 14, 18 and 6. These numbers have not been 
corrected for the errors due to absorption, etc., so that they give only an 
approximate estimate of the relative intensities of the emission lines. The 
lines 72 and 73, however, lie very close together, and the correction must 
be very small. Since the critical voltages for all the 7 lines are not quite 
the same, their relative intensities depend somewhat on the voltage 
