312 Royal histitutiun. 



plate a third class of experiments, namely, the determination of the 

 difference of gravity at the top and the bottom of a deep mine, by 

 pendulum experiments. Supposing the difference of gravity found, 

 its application to the determination of density (in the simplest case) 

 was thus explained. Conceive a spheroid, concentric with the ex- 

 ternal spheroid of the earth, to pass through the lower station in the 

 mine. It is easily shown that the attraction of the shell included 

 between these produces no effect whatever at the lower station, but 

 produces the same effect at the upper station as if all its matter 

 were collected at the earth's centre. Therefore, at the lower station 

 we have the attraction of the interior mass only : at the upper sta- 

 tion Ave have the attraction of the interior mass (though at a greater 

 distance from the attracted pendulum) and also the attraction of the 

 shell. It is plain that by making the proportion of these theoretical 

 attractions equal to the proportion actually observed by means of 

 the pendulum, we have the requisite elements for finding the pro- 

 portion of the shell's attraction to the internal mass's attraction, 

 and therefore the proportion of the matter in the shell to the 

 matter in the internal mass ; from which the proportion of density 

 is at once found. Moreover, it appeared probable, upon estimating 

 the errors to which observations are liable, that the resulting error 

 in the density, in this form of experiment, would be less than in the 

 others. 



Accordingly, in 1826, the Lecturer, with the assistance of his 

 friend Mr. Whewell (now Dr. Whewell), undertook a series of expe- 

 riments at the depth of nearly 1200 feet, in the Dolcoath mine, near 

 Camborne, in Cornwall. The comparison of the upper and lower 

 rocks (to which further allusion will be made) was soon found 

 to be the most serious difficulty. The personal labour was also 

 very great. They had, however, made a certain progress when, on 

 raising a part of the instruments, the straw packing took fire (the 

 origin of the fire is still unknown), and partly by burning and partly 

 by falling, the instruments were nearly destroyed. _^ 



In 1828 the same party, with the assistance of Mr. Sheepshanks 

 and other friends, repeated the experiment in the same place. After 

 mastering several difficulties, they were stopped by a slip of the 

 solid rock of the mine, -Which deranged the pumps and finally flooded 

 the lower station. 



The matter rested for nearly twenty-six years, the principal pro- 

 gress in the subjects related to it being the correction-to the compu- 

 tation of " buoyancy " of the pendulum, determined by Colonel 

 Sabine's experiments. But in the spring of 1854, the manipulation 

 of galvanic signals had become familiar to the Astronomer Royal 

 and the Assistants of the Greenwich Observatory, and it soon oc- 

 curred to him that one of the most annoying difficulties in the for- 

 mer experiment might be considered as being practically overcome, 

 inasmuch as the upper and lower clocks could be compared by simul- 

 taneous galvanic signals. Inquiries, made in the summer, induced 

 him to fix on the Harton Colliery near South Shields, where a re- 

 puted depth of 1260 feet could be obtained; and as soon as this 



