a34 AN ACCOUNT OF THE 
the reason. As I had neither the facilities nor the funds to allow of 
my conducting the operation in the same style, I saw that some check 
was required, to guard against any very great change ia the length of 
the rods. To compare them several times a day, would have been a 
means of delaying excessively the operation, especially as having no 
one I could depend on, to afford me any assistance, added to which, I 
had no materials of which to construct the standard rod, except wood, and 
then I had no means of guarding it against the effects of the weather. Itis 
true there was the chain, (and an invaluable standard of comparison it 
proved) but to have compared the rods with it daily, even once, not to say 
several times, would have caused so much delay, as must have deprived me 
of all hopes of finishing the work, within any reasonable period. ‘To lay 
off the length of the chain it was necessary to insert firmly into the ground, 
a draw and a weigh post, and this consumed much time: again without 
stands and coffers, it was the work of half a day to get the chain correctly : 
laid. It was indeed a consideration of these difficulties, that made me 
originally abandon the idea of using the chain in the measurement, and 
yet in practice, I found them much greater than I had imagined. As 
therefore it was quite out of the question, comparing the rods often with the 
chain, I thought of the following plan of detecting any changes in their 
length, arising from variations of temperature or humidity. 
12. Tue original idea of this plan was unexceptionable, and if it had 
been executed, would have stamped the measurement with every appear- 
ance of accuracy. Unfortunately however I was tempted to modify it, in 
consequence of some difficulties that occurred, and by this modification an 
