232 ILLINOIS STATE ACADEMY OF SCIENCE 



determine with considerable care whether an X-ray beam 

 has any effect on the resistance of an ordinary metallic 

 conductor. As tin foil was most convenient it was chosen 

 for a preliminary experiment. 



The foil was cut into the form of a grid so as to have 

 a total resistance of about 13.1 ohms, not including heavy 

 leads soldered to it. This grid was mounted with shellac 

 on a glass plate and enclosed in a light pasteboard box 

 for the sake of protection from air currents and rapid 

 temperature changes. The grid was placed in a position 

 about twenty centimeters from the target of a Coolidge 

 tube. The current for the tube was furnished by a large 

 Klingelfuss induction coil with a Wehnelt interrupter. 

 The equivalent spark gap was approximately six inches. 



The resistance measurements were made mth a Wheat- 

 stone bridge circuit arranged so as to be especially con- 

 venient for the detection and measurement of small 

 changes in resistance, essentially a bolometer. The slide 

 wire was shunted and also resistances w^ere inserted at 

 each end so that its equivalent length was about fifteen 

 hundred meters. The galvanometer was a special Leeds 

 and Northup instrument of high sensitivity and low re- 

 sistance. The bridge arms were so proportioned that the 

 galvanometer was critically damped. The use of a good 

 telescope and a scale three meters from the mirror aided 

 in securing high sensitivity. 



The sensitiveness of the arrangement could be deter- 

 mined in two ways, by the deflection resulting from' a 

 given shift in the balance point and by the deflection pro- 

 duced when a known change of resistance occurred in the 

 X-branch. As it was more convenient and more accurate, 

 the latter method was employed. In series with the tin 

 foil grid was placed a one ohm resistance which could be 

 shunted by a resistance of ten thousand ohms. The re- 

 sulting change in the resistance of the X-branch was very 

 nearly one ten thousandth of an ohm. It produced a 

 change of twelve millimeters in the galvanometer deflec- 

 tion. For resistance changes up to at least three thous- 

 andths of an ohm there was a linear relation between the 

 resistance increment and the corresponding galvano- 

 meter deflection. During all measurements the galvano- 



