84:6 Dr. C. V. Burton : An Experiment indicating 



suspension-chamber to the highest vacuum obtainable by 

 means of a Topler pump; but it was very difficult to ensure 

 that the whole was really air-tight; and an air-jet entering- 

 at a microscopical leak in some unknown position might well 

 become a source of serious uncertainty, so in the actual 

 observations the plate was surrounded by air at atmospheric 

 pressure. The suspension-chamber was formed of a mild 

 steel tube having an internal diameter of 3*2 cm., with walls 

 about 0'6 cm. thick, and of sufficient height to allow of the 

 use of a suspending fibre 30 cm. long. This tube was 

 hermetically sealed below, and was closed above by an 

 air-tight conical cap. It was provided with a thick window 

 of worked glass, and was fixed by screwing into a large 

 circular cast-iron plate, which in turn was adjustably bolted 

 to the cast-iron bed-plate of the apparatus ; the plate also 

 carried a cast-iron cell in which an achromatic object-glass 

 was geometrically seated. As an indication of the pre- 

 cautions taken against convection currents, it will suffice to 

 mention the several protecting shells which would be revealed 

 by a section in a horizontal plane through the suspended plate. 

 Proceeding outwards from the plate, we should encounter 

 first the sides of a box * measuring about 1*2 x 1'2 x 1*2 cm. r 

 and made of aluminium-foil, with a front of selected micro- 

 scope cover-slip; then the walls of the steel tube with worked 

 glass in front ; then a cylinder of sheet zinc (which had a 

 wooden cover); and finally the sides of a box made of 

 half-inch match-boarding. This last was supported from the 

 cement floor independently of the apparatus, and was closed 

 as completely as possible, only such small apertures being- 

 left as were determined by the optical requirements, and by 

 the necessity for keeping the woodwork out of actual contact 

 with the bedplate. 



9. The entire apparatus, including the vital parts of the 

 optical system, was firmly clamped upon the bedplate, and 

 this in turn was supported on three steel balls of 1*6 cm. 

 diameter, which were interposed between it and the seating. 

 Of these balls the first had three points of contact with the 

 seating and three with the underside of the bedplate ; the 

 second lay between a V-groove in the seating and a similar 

 groove in the underside of the bedplate, these grooves 

 pointing in the direction of the first ball ; while the third 

 ball lay between two horizontal plane surfaces, one belonging 

 to the seating and the other to the bedplate. Those parts of 

 the seating which contributed to this double " hole slot and 



* This box contained a trace of uranium oxide, to eliminate variable 

 electrostatic effects. 



