SECT. 4] 



403 



would soon reveal that scattering alone would not account for the total attenua- 

 tion of the original beam. We conclude, therefore, that the medium, in addition 

 to inducing a loss by scattering, also "absorbs" some of the beam's radiance. 

 a is apparently the sum of two generally independent terms : a term, s, which 

 refers to that part of the attenuation due to scattering of flux from the beam 

 without change in wavelength, and a term a which refers to the conversion of 

 radiant flux into other forms of energy some of which reappear as flux of 

 different wavelength than that of the original beam. Thus, we may write 

 a = a + s. 



Equations (10) and (11) supply the following alternate operational definitions 

 of a: 



I dNr K /Nr\ 



(12) 



b. Volume scattering function 



In the preceding discussion, during the attempt to estimate the amount of 

 radiant flux scattered out of the original beam, the experimental arrangement 

 shown in Fig. 6 was used : the Gershun tube G was directed at a fixed point p 



Fig. 6. Schematic diagram of optical system for determining volume scattering function, 

 aid). 



in the original beam. The tube was turned so that it successively looked at 

 point p in all directions 6 from the direction of the source, S. Thus 6 was varied 

 essentially from to 180°. For each orientation d, G was always kept at a small 

 fixed distance r' from p, and the corresponding field radiance Ni*{d) of the beam 

 was recorded. The length, l{d), of the path of sight through the beam for that 

 particular orientation was also noted. From this information, the flux scattered 

 out of the beam over each unit of path length can easily be computed. We now 



