682 
ME. AIET’S ACCOUNT OF THE COXSTEUCTION OF 
I proceed now to allude to a discordance wMcli was a source of great anxiety to 
Mr. Sheepshanks. 
In April 1855, Mr. Sheepshanks was engaged in measuiing the bar Cast Steel D. 
By comparisons with four iron bars (as is stated in the Table abore), whose results 
agreed very closely, the excess of Cast Steel D above Bronze 28 was found to be — 3'^-61. 
But a direct comparison of Cast Steel D with Bronze 28, immediately preceding, had 
given — 0‘’-46. This comparison was made at the temperature 45°-54, or 16°-46 below 
the standard temperature. A trifling error of expansion might account for part of the 
discordance ; and ordinary errors of observation might account for part. But in the 
opinion of Mr. Sheepshanks, though the whole discordance scarcely exceeded the effect 
of the thermometric expansion of Bronze 28 for 0°’3 Fahkenheit, it was impossible so 
to explain away the whole or a large part of it; and he was con-vTnced that Bronze 28 
had sensibly shortened. And so deeply and so painfully was this impression fixed in his 
mind, that he actually contemplated the rejection of all the results which had cost so 
many years of labour, and the commencing the work de novo. 
The first conjecture on the possible cause of the apparent change was, that Bronze 28 
while covered with gold-beater’s skin and cement (as in the earlier comparisons) might 
have been so far constrained by that covering that it could not shrink down to its 
natural length ; but that in the last comparisons with Cast Steel D, when that covering 
had been removed, it had contracted itself. In order to test this, ]Mr. Sheepshanks 
immediately made measui’es of Bronze 12, Bronze 29, Bronze A, and Bronze B, whose 
coverings had not been removed ; then he had the covering removed, and repeated the 
measures. The following is a comparison of these measures : — 
Excess above Bronze 28. 
d. 
Bronze 12, before cleaning, 1855 
April 26 
-f4-50 
after cleaning. 
May 1 
+4-77 
Bronze 29, before cleaning. 
May 5 
+ 3-70 
after cleaning. 
May 8 
+ 4-28 
Bronze A, before cleaning. 
May 9 
-2-36 
after cleaning. 
May 11 
-2-46 
June 5 to 
9 -2-51 
Bronze B, before cleaning. 
May 10 
+ 4-15 
after cleaning. 
May 12 
-f4-51 
May 31 
+ 3-57 
June land2-]-4‘59 
It seems sufficiently evident that nothing depends on the covering. 
Numerous observations were then made, partly by Mr. Sheepshanks himself and 
partly by a young assistant, on twenty-seven bars (including the four just mentioned) 
which had been previously well compared with Bronze 28. The result is exhibited in 
the following Table : — 
