PROFESSOR THOMSON ON THE ELECTRO-DYNAMIC QUALITIES OF METALS. 675 
several slips of copper cut away from each side of the conductor in the parts between 
the heater and the coolers, leaving the parts within these vessels unchanged. 
51. Several experiments were made on the conductor thus reduced successively to 
smaller and smaller numbers of slips ; but the results did not appear much more 
decided than they had been in the experiments on the unreduced conductor, until it 
was tried with all the slips but two cut away. Thus with four slips left, the following 
results were obtained : — 
52. Copper conductor reduced to four slips. 
Experiment VII. P'ebruary 1854. 
(Current six times eight minutes in each direction.) Temperatures and 
differences of temperature after eight minutes of current entering 
Augmentations of 
differences from 
middles to ends of 
By end next A. 
By end next B. 
periods. 
Periods. 
Ta- 
Tb- 
Ta-Tb=D. 
Ta- 
Tb- 
Ta-Tb=D'- 
D'-D. 
I. 
46-00 
45-24 
-76 
45-92 
45-27 
■65 
— •11 
II. 
45-82 
45-19 
-63 
46-01 
45-43 
^58 
— '05 
III. 
46-00 
45-50 
•50 
46-02 
45-54 
•48 
—•02 
IV. 
46-20 
45-77 
•43 
46-21 
45-76 
•45 
•02 
V. 
46-18 
45-80 
•38 
46-09 
45-71 
•38 
•00 
VI. 
46-09 
45-72 
•37 
46-05 
45-68 
•37 
•00 
Means for five periods.. 
46-058 
45-596 
-462 
46-076 
45-624 
•452 
— •010 
Diminution of difference during periods included *28 
Add average diminution per half-period 4- ‘028 
Effect due to reversal of current *018, 
in favour of Vitreous Electricity. 
The effect shown here is of the same kind as had been found in all the previous 
experiments, but was still too small to be very satisfactory. Some unknown cause 
made the difference T^ — Tg to diminish so much through the whole experiment as to 
overpower the apparent tendency of the current from B to A to increase it, and the 
abridged table has on this account a very unsatisfactory appearance as regards the 
conclusion drawn from it after the proper correction for their diminution is applied : 
but the full examination of the progress of variation in the course of the experiment 
shown in Table IV. below is much less unsatisfactory, and shows undoubtedly the 
true convective effect in copper. 
53. The following results, derived from the first experiment made on the conductor 
reduced to two slips, show a very marked increase in the effect, and make the result 
quite apparent even without the full analysis given below in Table V. 
4 u 
MDCCCLVI. 
