686 



PROCEEDINGS OF THE AMERICAN ACADEMY. 



by an automatic device after running three seconds. Though this 

 reduced the size of the deflections, it rendered less rapid the change 

 in condition of the tube. The deflections of the instrument under 

 these conditions appear, from considerations of the motion of the 

 suspended system under the forces involved, to be proportional to 

 the energy received during the exposure, just as in the case where 

 the greatest attainable deflection is used. 



A comparison of the mean curve for the solid screens with the curve 

 for the laminated screens should show the effect upon transmission, if 

 any, due to the surfaces of the screen. In Figure 5 the dots and the 

 curves drawn through them are from the data obtained with the solid 

 screens ; the crosses are points on the curve for the laminated screens. 

 The screens used in the experiment with aluminium were as follows : 



Solid screens : No. 1, 0.027 cm. 

 No. 2, 0.050 cm. 



Laminated screens : No. 1, 0.027 cm. 



No. 2, 0.054 cm. 

 No. 3, 0.081 cm., 



No. 3, 0.072 cm. 

 No. 4, 0.160 cm. 



1 layer. 



2 layers. 



3 layers. 



No. 4, 0.162 cm., 6 layers. 

 Those used in the experiment with copper were : 



No. 3, 0.021 cm. 



Solid screens : No. 1, 0.0044 cm. 

 No. 2, 0.014 cm. 



Laminated screens : No. 1, 0.0044 cm., 1 layer. 



No. 2, 0.013 cm., 3 layers. 

 No. 3, 0.023 cm., 5 layers. 



TABLE VI. 

 Summary of Table V. 



