“944 AN ACCOUNT OF THE 
I immediately afterwards compared this pair with the other, which had not 
-been ‘touched, by substituting them, alternately between two fixed points. 
‘The trial was satisfactory, and proved they had not altered their length by 
any sensible quantity. 
95. Tuts base was measured twice roughly, before commencing the 
correct measurement, and after fishing it. The operation, each time was 
performed with a Gunter’s chain of sixty-six feet, compared with a wooden 
rod, the length of which had been laid off from the brass scale. The 
length by these two measurements came out 21,766 and 21, yf 46. 'The true 
‘measurement as reduced to the level of the sea, and temp 69—~is 21754-8— 
| So. that the mean of the above two would come very near the truth. At 
alle events their near agreement with it shews, that no material error or 
omission had eon pomitoitted. 
_ 26. BErore deducing the real length of the line from the details 
given 1D. the accompanying paper, some thing must be said of the man- 
ner of deter mining the length of the rods. There were two methods, which 
presented themselves either to compare the four rods placed together with 
the chain, or to lay off twenty-five feet by means of the brass scale, 
on one of them, and compare the other three with it—as a check on the 
‘operation. { determined to try both methods and it is satisfactory to find 
that they agreed so nearly—-the difference between the two values thus 
‘ndépendently obtained, amounting only to eight feet, on a distance of four 
‘files. “As however Mr. 'Trovcnton had omitted to mention, either in 
