Vertical Force Magnetograph. 397 
to an arrestor J, which can be raised by a cam X worked by 
a handle Y. When the arrestor is raised the right-hand 
magnet rests in two V’s, while the left-hand magnet rests on 
a plane surface. 
The instrument is closed in with a wooden case, and a ther- 
mometer passes through the upper surtace of this case, the 
bulb being in the seieiihourhaod of the magnets. A still 
drawer is fitted at the lower part of the base ‘which serves to 
contain some desiccating agent when the instrument is used 
in a damp situation. 
An instrument of the above pattern was, through th: 
kindness of the Director (Dr. R. T. Glazebrook), tested at 
the National Physical Laboratory, and was found to work 
satisfactorily. Owing to disturbances produced by electrical 
railways it was impossible to make the gio compen- 
sation at South Kensington ; and Mr. F. £. Smith, of the 
National Physical Jaboratory, kindly undertook to perform 
the adjustment after the instrument was removed to Bushey 
House. He performed this adjustment with such skill, that 
with a change of temperature of 7° C. he was unable to 
detect any variation in the scale-reading (1 mm.=1*6y) due 
to temperature change. When making these temperature 
adjustments, the trace obtained from a vertical-force magneto- 
graph which was kept at a constant temperature was used to 
eliminate changes in V. I should like here to express my 
thanks to Mr. Smith for the trouble he took, and my appre- 
ciation of the skill with which he handled what is necessarily 
a very fragile and delicate instrument. In fig. 7 is given 
a reproduction of a portion of a trace obtained with the instru- 
ment when at Bushey. In the original 1 cm. corresponded 
to 3 minutes of time and 8y~* respectively. The recording- 
drum was at a distance of 170 cms. from the instrument. 
The following hints as to making and adjusting the move- 
able system may be of assistance to any one attempting to 
make one of these instruments. A slab of fused silica is cut 
by means of a lapidaries’ wheel, armed with diamond-dust, 
and roughly ground to size with carborundum on a metal 
plate +. Two small tags of silica are then attached to the 
sides of this slab by fusion withan oxyhydrogen flame. These 
tags serve to support the stirrups which carry the magnets, 
and to their ends the fibres will be fused afterwards. The sur- 
faces of the slab and of the tags are ground plane and approxi- 
mately parallel, and then one » surface is ground and polished 
* ]y='00001 c.G6.s. unit. 
T The methods of working such quartz mirrors and producing the 
platinum reflecting surface are described inthe paper referred to previously. 
Phil. Mag. 8. 6. Vol. 7. No. 40. April 1904. 2H 
