16 O. N. Rood on thermo-electric currents by percussion. 
To avoid the effects of conduction external to the couple, a 
number of insulating substances were tried, which gave more or 
less constant results. 
1. Thin card board or plates of mica placed above and below 
the aganle gave irregular resu 
o thicknesses of dried bladder, paiannd above and below 
the ‘saoke gave somewhat better results 
3. Four omeeaierigg of heavy woven silk were also used for 
the same pu 
4. The best ae were, however, obtained by using heavy 
woven silk, which was spread over with a coating of yellow 
wax, and then wrapt around the sa at the juncture. This 
insulating seen after being used for some time so as to 
, gave results which were about as constant as 
could be ie under the circumstances 
low are results obtained in these several WARE 
Taste 2.— With two skins above and below. 
Distance fallen, lin. 2 in. 3 in. 4 in. 
3 15? 2°29 3°0° 58° 
1-4 2:3 4°3 
1:8 2°5 3-0 4-4 
First unreduced 1:2 3-5 3°4 5-9 
deviations, ; 1-4 30 35 6-0 
1-2 35 3°7 5-0 
11 24 41 58 
11 2-2 B°4 49 
Average, 1:3 2-7 3°55 52 
In this and in all the other tables, the order of the experi- 
ments was across the page, from left to right, and not down the 
single columns. 
e No. 8, contains results obtained with four layers of 
plain silk above and below the couple. 
Tase 3. 
Distances fallen, 1 in. 2in. 3 in. 4in. 
6F 34° 37° 61° 
“4 3°9% 54 
16 3°5 41 6:0 
First unreduced is py 5 42 60 
deviations, 15 26 43 60 
5 2 50 65 
“4 26 47 60 
“4 2 43 69 
Average, 16 30 45 60 
The results given in tables 2 and 3, were obtained by using a 
comes wire of German silver and iron with a diameter of 9 
of a millimeter; the juncture was bound with a little fine iron 
wire and soldered. This form of couple was found to lose its 
shape by the repeated blows; it also finally cut the insulating 
substance, so that in all the following experiments plates of the 
| 
