388 



MAGNETIC OBSERVATORY AT 



row limits. A screw at the top of the frame serves to adjust the 

 magnet to the proper height. The circular top plate is graduated 

 on its edge, and can be revolved about a central collar, by which 

 means the suspension skein can be turned through any desired angle. 

 In order to take out all twist, when the instrument is first mounted 

 a bar of lead of the same weight as the magnet is suspended in its 

 place, and after it has found its position of rest, the upper plate or 

 "torsion circle" is turned until the bar comes to rest in the magnetic 

 meridian. After suspending the magnet the effect of torsion of the 

 suspension skein on the scale value is determined by turning the 

 torsion circle through 90° on either side of its proper position, find 

 observing the change of direction produced on the magnet. The 

 ratio of this deflection to 90° is the proportion in which all observed 

 variations recpiire to be increased, in order to correct for the resist- 

 ance of the suspension skein to torsion. 



SOURCE OP LIGHT AND METHOD OF REGISTRATION. 



Gas is brought to the magnetic observatory in a leaden pipe from 

 the main which supplies the Smithsonian Institution. Before reaching 

 the burners it is made to pass over naptha exposed in a suitable 

 vessel, by which means it gives a more white and brilliant light. 

 The shape of the burners is that known as fishtail, and they are so 

 placed that the plane of the burner makes a small angle with the 

 direction to the mirror, in order to allow light from the whole width 

 of the flame to reach the mirror through the narrow aperture of the 

 chimney which surmounts it. This chimney is an oblong blackened 

 copper tube, having a small opening about one-tenth of an inch wide 

 and half an inch high on the side towards the mirror, and close to 

 this opening, on the outside of the chimney, is a screen, with a 

 narrow rectangular slit, adjustable in width by a sliding plate. A 

 good working width of this slit is found to be a little over one hund- 

 redth part of an inch by four-tenths of an inch in height, which forms 

 the beam of light, a magnified image of which is thrown upon the 

 Just before reaching the cylinder this beam is 

 o received upon a pair of 



record cylinder. 



cylindrical lenses, seen in 

 section in fig. 3, by which 

 it is refracted to a focus 

 of about one -twentieth 

 of an inch diameter, 

 which makes the pho- 

 tographic trace. By 

 this ingenious arrange- 

 ment a sufficient quan- 

 tity of light is concen- 

 trated to produce a very 

 good trace. The lenses 

 extend the whole length 

 ! he cvlinder, and are 



mounted in a frame, which can be removed and replaced in position 



e very time the paper on the cylinder is changed. 



