Chemistry and Physics. 393 
acts like a conducted motion, the total motion in the sector is 
oe 1 But if the action of convective had been greater 
a 
or less than that of conducted motion, the excess, in one or the 
other direction, must have been shown on the sector. 
€ experiments showed that, when the sector was small (} of 
the circumference), the small difference between the convection 1 
e conduction 4 in general could not (or at least not with 
certainty) be detected, that therefore, with approximate equality 
of convection and conduction, the electro-dynamic effect of the one 
sensibly neutralized that of the other. 
en, however, the sector took in half of the circumference, the 
here assumed current could be observed even in the free portion 
of the disk, though the amount was too small for safe measure- 
ment 
The observed electro-dynamic action being so little in the fore- 
golng experiments, in which the disk was electrified and covered 
i its whole extent by the induced plates, theoretical calculation of 
the amount of the action from the known absolute values of the 
electro-dynamic constants promised only approximately accordant 
values, Nevertheless it was carried out by Mr. Rowland. 
€ proportion in which the action of the earth’s magnetism upon 
the pair of astatic needles was diminished was ascertained by find- 
Ing the oscillation-period, first with the needles equally directed, 
and then arranged astatically. 
law of spark-leneth given by Sir William Thomson, which in this 
case appeared sufficiently accurate. Before and after each experi- 
Ment, a smaller jar was charged from the battery of nine large 
ones containing the store of electricity, and on it the length of the 
Spark was determined. 
sane velocity of the rotation was regulated by the position of the 
lls of a centrifugal governor, which was applied to one of the 
Tevo ving : 
lise the calculation of the distribution of the electricity on the 
‘*, and its electro-magnetic directing-force, the surplus charge 
Value wh; margin of the disk was reckoned according to the 
infi eel holds for infinitely thin disks, and considered as an 
witely thin thread concentrated at the margin—a proceeding 
of i wasn both ways only approximately correct, but, in presence 
€ minuteness of this portion, was sufficient. 
1 
® iniluence exerted upon the upper needle was about 3; of 
r 
ciate uh, Which was for some time attached to the most rapidly 
axis, 
m was put equal to 0°182 (using the centimeter, gram, and 
Second as units): the electro-dynamic constant was put by Mr. 
