ELECTRIC DISCHARGE WITH THE CHLORIDE OP SILVER BATTERY. 485 
natural direction. It would appear, therefore, that the natural discharge in the 
intervening space in the outer tube is, as it were, replaced by a non-luminous one. 
When the batteries were kept on for about twenty seconds, or so, these intervening 
strata gradually diminished, and at last disappeared entirely, as shown in Plate 35, 
fig. 4, where they are replaced by a non-luminous space ; but it is very evident, from 
the shape of the strata where visible, that two discharges continued to take place in 
the direction proper to each from both sets of terminals; the last stratum of the small 
tube on the positive side curving upwards into the broad space. 
On keeping the batteries in connexion for a further period another remarkable 
change took place, and the discharge in the larger tube appeared to gain the mastery, 
its strata in the interval between and above the open ends of the small tubes 
reappearing with a forward movement one by one at the positive side, the convexities 
being turned towards the negative, which is the natural direction. The strata in the 
small tubes did not continue beyond their extremities, and a dark space intervened. 
These phenomena are shown in Plate 35, fig. 5. Keeping on the discharge still longer 
the strata in the small tube disappeared entirely, as shown in Plate 35, fig. 6. 
Ultimately, by continuing the discharge, both sets of the strata disappeared, as if 
one discharge neutralized the other, but these could be reproduced by breaking and 
remaking the contact with the batteries. 
It is very probable when two discharges take place in the same tube in opposite 
directions that the column of gas becomes divided longitudinally into two layers, one 
of which conveys the discharge of one battery, while the other conveys that of the 
other battery. 
Two separate currents crossing each other at right angles. 
In order to ascertain the effect of two discharges from two distinct batteries cross¬ 
ing each other at right angles, we prepared a tube, No. 344, consisting of two tubes 
joined at right angles to each other, each tube being 33 inches (83‘8 centims.) long 
and 1^ inches (2’86 centims.) in diameter. The central portion is shown in fig. 11. 
Fig. 11. 
TV 
O 
1 DP DD 
TV 
In the first place the discharge of a battery of 3600 cells was sent through one of 
the tubes containing air at a pressure 0'428 m.m., 563 M, causing a current of 0'01507 
ampere, and producing the stratification which is represented in Plate 35, fig. 10 ; 
MDCCOLXXXIir. 3 R 
