266 ANNUAL EEPORT SMITHSONIAN INSTITUTION, 193 3 



strength is greater than that of air, then the voltage and the current 

 both increase proportionately and the power output increases as the 

 square of the break-down strength. The two media most convenient 

 are either some gas such as air at high pressure, or a vacuum. With 

 gas at high pressure, the break-down strength is approximately pro- 

 portional to the pressure, so that the operation of a generator in a 

 tank of gas at 30 atmospheres pressure should give 30 times the 

 voltage, 30 times the current, and 900 times the power of the same 

 device mounted in the open air. It is relatively easy to build a con- 

 tainer for compressed gases and to mount a generator in it, and this, 

 in fact, has been done by Dr. Barton at Princeton, originally with 

 the collaboration of Dr. Van de Graaff. 



By far the most intriguing possibilities of this generator are found 

 in its vacuum embodiment, because a high vacuum is the best of all 

 insulators since it offers no " windage " resistance to the motion of 

 the belt, and because many of the api)lications of the high voltage 

 will themselves be in vacuum discharge tubes which can be built 

 right into the generating system. 



Such a generator has been designed and built. It is still in the 

 experimental stage, but various complicating factors have one by one 

 been overcome. Experience to date indicates that there is in sight 

 no insurmountable obstacle to the construction of generators which 

 may even reach considerably higher voltages than the generator at 

 Kound Hill. 



In conclusion, it is interesting to know that two Van de Graaff 

 generators have been built and operated in Washington under the 

 direction of Dr. Merle A. Tuve, of the Department of Terrestrial 

 Magnetism of the Carnegie Institution. One of these has been 

 actually used for experiments on atomic disintegration and the other 

 instrument, a larger one developing upward of 2 million volts, is 

 awaiting a suitable housing, now under construction, for its satis- 

 factory operation. Dr. Tuve in Washington, Dr. Coolidge, of the 

 General Electric Co., and Dr. Slack, of the Westinghouse Co., all 

 of whom have built and experimented with Van de Graaff generators 

 subsequent to the demonstration of Van de Graaff's first air-operated 

 instrument in the summer of 1931, have been very helpful in report- 

 ing their experiences with the generators. 



In conclusion, it may fairly be said that this new type of generator 

 as an electrical instrument has already been highly successful and 

 shows promise of very considerable further development. It remains 

 to be seen whether the necessity which was the mother of this inven- 

 tion — namely, the desire for high-speed particles for the study of 

 atomic nuclei — will lead to important new knowledge of atomic 

 structure with the aid of this device. Several good men are begin- 



