July 4, 1919] 



SCIENCE 



23 



records? Evidently there was no attempt at 

 sound ranging. 



The typographical work on the book is ex- 

 cellent except that the photographs, charts 

 and diagrams are poorly lettered and not up 

 to the rest of the book. 



A. M. 



SPECIAL ARTICLES 



VARIATIONS IN THE ELECTRICAL POTENTIAL 

 OF THE EARTH 



At a meeting of the Academy of Science of 

 St. Louis, held on March 17, the writer pre- 

 sented diagrams representing variations in 

 gravitational attraction between the masses of 

 the Cavendish apparatus in the second story 

 of the physics building at Washington Uni- 

 versity. 



; This apparatus is composed of a shield in 

 which the smaller masses are suspended on a 

 bi-filar suspension of silk fibers. The top, bot- 

 tom and ends of this shield are of wood, cov- 

 ered within and without with tin-foil. The 

 sides are of sheet metal, clamped to the wood 

 frame by bars of wood and the joints sealed 

 with wax. The wood clamps are covered by 

 tin-foil. The whole is then surrounded by two 

 end caps of metal which meet at the middle of 

 the shield and are sealed together with tin-foil. 

 The position of the suspended masses is deter- 

 mined by a telescope and scale in the usual 

 manner. The mirror is observed through a 

 narrow slit in the two metal screens which 

 surround the suspended masses, and which is 

 closed by a strip of glass sealed to the inner 

 sheet of metal. The suspended masses are elec- 

 trically charged by means of a wire armed 

 with a pin which is thrust through a glass 

 tube which is passed through a small opening 

 in the end of the shield. When connected 

 with the electrical machine in an adjoining 

 room the air within the shield and the sus- 

 pended masses were charged. This operation 

 was made to come about gradually by having 

 a gap armed with pins in the line leading to 

 the machine. In some cases the suspended 

 masses would swing into contact with the 

 metal sheets forming the sides of the screen. 

 It was arranged that they should be deflected 

 towards the large masses. It was found that 



ou withdrawing the glass tube and pin and 

 closing the opening in the screen with tin-foil, 

 the small masses could be liberated with an 

 initial velocity approaching zero, by connecting 

 the large masses and screen directly with the 

 machine terminal, eliminating all gaps in the 

 line. The impression thus created was that 

 gravitational attraction was thus diminished 

 until the torsional effect of the bi-filar suspen- 

 sion could detach the smaU masses from the 

 screen, to which they were held by an elec- 

 trical attraction. 



After the suspended masses had come to rest 

 in the center of the screen, which was usually 

 on the following day, the large masses were 

 directly connected with a large copper rod on 

 the outer wall of the building, which served as 

 lightning protection for the building. This 

 rod was the ground connection for a steel tower 

 on the roof of the building, which formerly 

 was part of a system for wireless telegraphy. 

 The top of this tower is 100 feet above the 

 ground. This tower and the earth replaced the 

 electrical machine, in the electrification of the 

 large masses. 



On clear days when there was practically no 

 wind, the gravitational attraction of the large 

 masses for the suspended masses has some- 

 times been diminished, until it has apparently 

 become a repulsion. All artificial heat was 

 cut off from the room, so that its temperature 

 increased during the day not more than two or 

 three degrees centigrade. The temperature of 

 the air in contact with the large masses was 

 under constant observation, the recording be- 

 ing made by means of a telescope. The tem- 

 perature could be read accurately to tenths of 

 a degree Centigrade, and hundredths of a de- 

 gree could be estimated with fair precision. 



When the masses were not in connection 

 with the lightning rod, the rise in temperature 

 during the day caused a very slow increase in 

 the reading which determined the position of 

 the suspended masses. This change was due 

 to convection currents within the shield sur- 

 rounding those masses. 



These convection effects have been very 

 carefully examined. They are distinctly ap- 

 preciable when the temperature of the room 



