326 Rev. Mr Scoresby on the Uniform Permeability of 



Now these experiments, with ordinary instruments, go far 

 enough to prove that the method here suggested for determining 

 distances, otherwise indeterminable, is capable of a degree of ac- 

 curacy sufficient to render the method practically useful in min- 

 ing operations ; for we find, that, with a small pocket-compass, 

 and two sets of bar-magnets (the largest only 2 feet in length), 

 distances amounting to above 8 feet are capable of being deter- 

 mined to within 3 inches, or only ^^\}cv of the whole ; whilst in 

 smaller intervals the proportion of error was often not more 

 than j^jd to ij^th of the whole distance. 



A single experiment, however, performed with the larger 

 pair of magnets, and one of Captain Kater's azimuth or survey- 

 ing compasses, will be sufficient to shew that this method of de- 

 termining the thickness of solid substances, is capable of a much 

 greater degree of accuracy. At the station in which the Liver- 

 pool and Manchester Railway Company have their fixed en- 

 gines for drawing carriages, &c. through the tunnels, two exca- 

 vations are cut in a solid freestone rock for the engine boilers. A 

 trial was made to measure the thickness of the septum of freestone 

 between the boilers, as a test of the degree of accuracy of which 

 this method is capable. The compass being placed on one side, two 

 feet from the outer edge of the septum or wall, and the two mag- 

 nets (B), on the other (bearing east from the compass), the 

 difference of the deviations produced by the opposite poles of 

 the magnets was found to be 15° 20'. The magnets were then 

 brought round to the west side of the compass, when equal de- 

 viations were obtained at the distance of 3 feet 5 inches from 

 the centre of the compass. On measuring the front of the wall, 

 its breadth was found to be 3 feet 1 inch, which, added to 3| 

 inches, the distance of the centre of the compass from the wall, 

 gave the total distance 3 feet 4f inches, indicating an error 

 of |th of an inch. Being a good deal perplexed with the amount 

 of this error, which, on repeating the experiment, was still the 

 same, it occurred tome, that the thickness of the septum, which 

 appeared to be so uniform, might possibly not be the same. To 

 ascertain this, the lines formed by both sides of the wall were 

 projected forward to the distance of 8 feet, when, instead of 

 being parallel, they were found to approximate about 4 inches, 

 or an inch in every two feet. Hence I found that the wall two 



