374 SCHAEFFER. 



results are at hand, however, which show the variation of such an 

 effect with wave-length. All that can be stated is that the photo- 

 electric effect on dust may cause part of the diminution of intensity 

 and the deviation from the scattering curve as recorded in Figure 1. 



Conclusion. — It seems that (1) the lower atmosphere is practically 

 free from ozone, (2) molecular scattering alone will not account for the 

 attenuation of the shorter wave-lengths from distant terrestrial 

 sources, (3) absorption by oxygen and ionization of the air may ex- 

 plain the deviation from the curve for molecular scattering, (4) prob- 

 ably the isothermal layer of our atmosphere contains the ozone which 

 sets the limit of the solar spectrum in the ultra-violet. 



My thanks are due Professor Lyman for many suggestions and for 

 his aid in securing suitable locations for conducting this work, and Dr. 

 Stetson for the use of his apparatus in measuring the opacities. 



Jefferson Laboratory, 



Cambridge, February, 1922. 



