g of
the circuit would produce no action. The motion of F produces a
_mechanical_ motion of the spring of the vibrator, which latter acts for
a few instants and produces a certain number of contacts that give rise
to an equal number of sparks. Owing to this arrangement, the expenditure
of electric energy required by each lighting is limited; and, an another
hand, the vibrator, which would be incapable of operating if it had to
be set in motion by the direct current from the pile, can be actuated
_mechanically_. As the motion of the vibrator is derived from the hand
of the operator, and not from the pile, it will be comprehended that the
latter can, everything being equal, produce a larger number of lightings
than an ordinary bobbin and vibrator.

[Illustration: Fig. 3.--INCANDESCENT GAS LIGHTER.]

Dr. Naret's _Fiat Lux_ (Fig. 3) is simpler in its operation, and cheaper
of application, since it takes its current from the ordinary piles that
supply domestic call-bells. It consists essentially of a fine platinum
wire supported by a tilting device in connection with the two poles of a
pile composed of three Leclanche elements. Upon exerting a vertical
pressure on the button placed to the left of the apparatus, either
directly or by means of a cord, we at the same time turn the cock and
cause the platinum spiral to approach, and the latter then becomes
incandescent as a consequence of the closing of the circuit of the pile.
After the burner is lighted it is only necessary to leave the apparatus
to itself. The cock remains open, the spiral recedes from the burner,
the circuit opens anew, and the burner remains lighted until the gas is
turned off. This device, then, is particularly appropriate in all cases
where there is a pressing need of light, for a single maneuver suffices
to open the cock and effect a lighting of the burner.--_La Nature_.

* * * * *

DISTRIBUTION OF HEAT WHICH IS DEVELOPED BY FORGING.

On the 8th of June. 1874, Tresca presented to the French Academy some
considerations respecting the distribution of heat in forging a bar of
platinum, and stated the principal reasons which rendered that metal
especially suitable for the purpose. He subsequently experimented, in a
similar way, with other metals, and finally adopted Senarmont's method
for the study of conductibility. A steel or copper bar was carefully
polished on its lateral faces, and the polished portion covered