189 
weight, it evolved 2,947 units, and when it worked rapidly, 
raising no weight, it showed 4,769 units. In that case, 
therefore, the effect was exactly the reverse of what Dr. 
Joule supposes. 
3. I have taken Dr. Joule’s own experiments described in 
the “Philosophical Magazine,” vol. XXVIII., and have 
calculated, on the data he gives, what weight ought to 
have been raised in each experiment. I subjoin a table 
of what should have been raised and what actually Ava,s 
raised : — 
Exp. 
Foot pounds calculated. 
Actual. 
I. 
19,377 
21,100 
II. 
18,628 
17,820 
III. 
11,533 
8,800 
IV. 
11,232 
9,000 
V. 
17,416 
10,031 
VI. 
18,033 
12,673 
In order that 
any mistake I may 
have made 
detected, I subjoin the calculations themselves in an 
appendix. The numbers are calculated on the supposition 
that the resistance of the battery could be neglected. If 
Dr. Joule in those experiments acted on the erroneous prin- 
ciple, into which he was misled by Jacobi, that the resistance 
of the wire should be equal to that of the battery, each of 
the above calculated numbers should be reduced by one 
half, which in every case would make the work actually 
done considerably more than the calculation, in some cases 
more than double. 
4. Let me next observe that if, instead of a wire 389 feet 
long and iVth of an inch in diameter, Dr. Joule had taken 
one which was half the length and half the section, all his 
figures would apply equally well; or, again, a wire of 
German silver a quarter the length and with four times the 
resistance; or, to take an extreme case, a wetted cord 
enclosed in an insulating tube Tooth part the length, and 
with 100 times the resistance. 
