ELECTRIC PHANOMENA OF THE ATMOSPHERE. 387 
the other has remained in the upper part, repelled by that of 
the globe. When the instrument is afterwards lowered, the globe 
acting with greater force on the upper end of the rod, forces the 
resinous electricity which had been accumulated there to distri- 
bute itself towards the inferior end, where the action is increased 
in a less proportion. It would seem at first that it is the contrary 
that should take place; that the resinous tension of the globe 
repelling the electricity of the same name should force it to 
escape and retain the other. This is what would take place if 
the rod were made to communicate with another body; but its 
communication with the globe itself, which is resinous, retains 
repelled that of the same name, and neutralizes the vitreous por- 
tion which is free. 
23. When the rod q is touched with an insulated proof plane, 
it does not show a preference for either electricity ; it carries off 
that which is repelled by the globe and not that which it retains 
by its induction. It is then the resinous tension which dimi- 
nishes and the vitreous tension which augments in the rod, which 
is contrary to the effect at the first contact, and which shows its 
opposition when the electroscope is lowered. 
In the first case, the resinous electricity, retained less at the 
upper extremity, distributes itself over the rod, and the leaves 
diverge with the appropriate sign ; in the second case, the leaves 
which were very divergent approach so near that they come into 
contact. 
24. This result may be produced as often as desired, and the 
electroscope may be neutralized at different heights, so that the 
instrument may give contrary signs at a given height. I suppose 
that the electroscope has been neutralized, the ball ¢ being even 
with the opening B; if it be raised three centimeters, the leaves 
will diverge vitreously: at this height, if the rod be made to 
communicate with the globe, the positive electricity will neu- 
tralize it in the globe and the leaves will fall to zero. If the 
instrument be raised three more centimeters the leaves will again 
diverge vitreously; and if we again establish the communica- 
tion of the rod with the resinous globe, this vitreous electricity 
will be there neutralized, as in the preceding case, and the resin- 
ous alone will remain repelled at the furthest extremity. If 
we then lower the instrument the three last centimeters, to 
replace it in the zone of its second equilibrium, and where its 
first divergence has been vitreous, the leaves then diverge re- 
