ON THE KEW OBSERVATORY. 361 



were gradually (and bodily) approaching the zero-line. It was therefore 

 examined by the maker of the magnet himself, who could see nothing in the 

 disposition of the other parts to prevent the proper action of the magnet. 

 The screen for preventing the bad effect of heat from the lamp was applied 

 (as to the other force instrument), and the slit in the moveable shield {¥), 

 having been found too narrow, was widened. 



On the 31st it underwent adjustments and examinations which we hoped 

 would be final. The radius was found to be 11 '93 inches (as on the 24th 

 inst.). The magnifying power of the lenses was 3-78 times, as ascer- 

 tained by the method described above. The arc value of 1 inch of the 

 ordinate scale was consequently 76''23. The time of vibration (viz. 17*9 

 seconds) in a horizontal plane, at a temperature of 52°, was determined by 

 suspending it, protected from currents of air, from a solid support by a small 

 silken thread, and by using a microscope with cross wires, to view a mark 

 on the shield (6^). Its time of vibration in a vertical plane, viz. 20 seconds, 

 at a temperature of 53°, was determined, by observing the oscillations of the 

 I'mag-e of the slit at the mouth-piece (E) and the chronometer. The time of 

 transit of a point of the plate over the mouth was found to-be about 1^ minute. 



On April 1 at m. it was brought into activity for the " trials ;" but the curve 

 soon began to exhibit great dislocations, and also the former tendency to ap- 

 proach the zero line. A condensing lens was afterwards mounted in front of the 

 camera(A),in order that the lamp might be removed to a greater distance from 

 it; but dislocations, &c. in the curve gave evidence of mechanical vibration 

 having occurred at the times of putting the plates into the slider-case (F) 

 and removing and replacing the lamp, i. e. at sunrise and set, and we 

 believed it possible that some cause of unsteadiness might still exist in the 

 mountings, particularly in the brass pillars (Q). These were therefore dis- 

 carded, ybr the present, and the intermediate corbel was very firmly planted 

 and cemented upon the plinth of the quadrant-wall to supply their place. 

 The whole stone and marble-work was rendered as secure against vibra- 

 tion as an experienced mason could make it. The magnet had altered its 

 time of vibration in the vertical plane to 1 seconds, and various changes of 

 this kind have since occurred (as stated). 



On the 9th the time of transit of a point of the plate over the mouth was 

 I5 minute. 



On May 2, the larger thermometer having been substituted for a smaller, 

 it was deemed ready to resume its functions ; but, on the next day, disloca- 

 tions, &c. corresponding with the usual times were shown by the curve ; 

 thus rendering it highly improbable that instability of the four very stout pil- 

 lars (Q) or of any fittings had produced the bad effect. 



On the 5th this conclusion was confirmed by discovering that the disloca- 

 tions were occasioned, principally, by the removal of the soft iron bars of the 

 neighbouring window-shutters, from a vertical to a horizontal position, and 

 vice versa at the above-mentioned periods (a precisely similar case to one which 

 had long since occurred relatively to the declination instrument, and had been 

 forgotten). The bars were therefore now (as was one bar formerly) discarded. 

 After this time the great dislocations ceased ; but a few instances have since 

 arisen of extremely sudden variations in the curves of all the magnets 

 simultaneously, which might have been mistaken for mechanical disturbances, 

 if they had not been simultaneous ; and some other small dislocations of these 

 vertical. force curves evince unnatural motions of the magnet. 



On June 4 it had altered its time of vertical vibration. 



