Notices respecting New Booh. 291 



points had been fixed, was carefully compared with a standard iron 

 bar of 10 feet, and the difference ascertained at the temperature of 

 62° Fahrenheit's thermometer. The unit of the measure, therefore, 

 so far at least as it depends on the measuring bars, was this 1 0-foot 

 iron bar. For the same reason it was necessary to determine the 

 exact distance between the outer foci of the microscopes, in each of 

 the seven sets of microscopic apparatus, in terms of a certain stand- 

 ard. The standard used for this purpose was a brass scale, on which 

 the fine dots were marked at a distance of 6 inches. In this pro- 

 ceeding we find another objection to the method. The part of the 

 base to which the bars were applied, that is to say |-^ of the whole, 

 is referred to a standard iron bar of 10 feet, and the remaining 2^ 

 to a brass standard scale of 6 inches. In order, therefore, to obtain 

 the length in terms of a single standard, it becomes necessary to 

 determine the exact relation of the 6-inch scale to the 10-foot bar, 

 for at least the whole range of temperature within which the mea- 

 suring apparatus was compared with the respective standards. Hence 

 it is necessary to determine, not only the exact relation of two scales 

 of very unequal lengths at a given temperature, but also their rates 

 of expansion. The circumstance of the two standards being con- 

 structed of different metals seems an unnecessary complication. 



The site selected for the measurement was on the eastern border 

 of Lough Foyle, in the county of Londonderry. It was proposed 

 to commence the measurement from the strand, near Mount Sandy, 

 at the entrance of the Lough, but the rugged and broken nature of 

 the ground, composed of low sand-hills, made it necessary to begin 

 about two miles to the south of the intended limit. From the 

 point which was selected the measurement was carried southward to 

 a small rising ground called Sheep Hill, about a quarter of a mile 

 south-east of the church of Ballykelly. The distance between the north 

 end and Mount Sandy was afterwards determined by triangulatiou. 

 The mean height of the ground above the sea at high-water was 

 about 18 feet. 



The whole length of the base, to which the measuring appa- 

 ratus was applied, is 41640'8873 feet, or nearly 7-|-^ miles. The 

 extension from the north end to a station near Mount Sandy is 

 1 15.59'8270 feet, or nearly 2^ miles. This last number is the mean 

 of the results of the triangulatiou computed in eight different ways. 

 It is probably just as accurate as that which expresses the measured 

 part of the base ; but admitting it to be so, we confess our inability 

 to perceive what particular advantage was proposed to be gained by 

 the prolongation. It gives no additional security whatever for the 

 accuracy of the computed distances, which depend solely and abso- 

 lutt'ly on the portion actually measured. If triangulation is to be 

 admitted, the side of any triangle, determined with equal precision, 

 may just as well be assumed as the base of the survey. 



For tlic purpose of verification the base was divided into seven 

 sections, and after two of them had been measured by the appa- 

 ratus, tiie length of the second section was computed from the first 

 by a triangulation. All the other parts were verified in a similar 



U2 



