PHYSICS AND NATURAL HISTORY OF GENEVA. 265 
influence of atmospheric agents in order to explain to a great extent the varia- 
tions of terrestrial magnetism. very rupture of meteorological equilibrium 
producing a condensation of watery vapor would produce a rupture of elec- 
trical equilibrium. This equilibrium cannot be re-established except by cur- 
rents of the surface, currents which must act on the magnetic needle. Doubt- 
less the mass of the sun exerts a direct action on terrestrial magnetism, but 
M. de la Rive thinks that this action has been much exaggerated. It would 
not surprise him if in the magnetometer theregwere eventually found an instru- 
ment of meteorology at least as delicate as the barometer. 
By the same colleague an account was given of the experiments relative to 
electrical cables, which he had witnessed in England. It appears that the ill 
success of the transatlantic cable is chiefly to be ascribed to the defective 
setting of the soldering, and also to the circumstance that the cable was laid 
in such a manner as was calculated to produce ruptures in the isolating en- 
velope. Nor is the fact to be overlooked, which has been proved by direct 
experiments, that pressure increases the conductibility of the envelope of gutta- 
percha, which is not so isolating as has been generally supposed. During an 
excursion to the neighborhood of Dover, M. de la Rive took oceasion to ob- 
serve the application of electrical illumination to light-houses. The electricity 
is not generated by a battery, but by magnets of steel arranged on a circum- 
ference before which pass points of soft iron surrounded by a coil and placed 
on the periphery of a wheel. ‘To this wheel motion is communicated by a 
small steam-engine. The cost of the apparatus once defrayed, and this cost 
is certainly considerable, the daily expenses are less than those of ordinary 
light-houses. We learn from a letter addressed to our colleague by M. Bee- 
querel, who has since studied the subject, that the light obtained by the above 
means is very constant. A machine of one and a half horse power, consuming 
six kilogrammes of coke per hour, suffices to produce currents which, issuing 
between two retorts of charcoal, give a light equal to 300 wax candles or 7g 
carcel lamps. 
Professor Wartmann reported to the society the researches of M. Magnus 
on the conductibility of gases. When a metallic wire is heated to redness by 
the current of a battery, it is found that the duration and intensity of the heat 
vary with the cireumambient gases; hydrogen, for instance, conducting heat 
as a metal would do. M. Wartmann gave an account also of a memoir of M. 
Rike, of Leyden, designed to explain the non-instantaneousness of the propa- 
gation of the electric fluid in conductors. He compares the propagation of 
electricity to the efflux of water or of elastic fluids under certain determinate 
conditions. ‘T’o conclude what relates to the principal reports concerning elee- 
tricity, I may here mention that M. L. Soret presented the model of a new 
battery, constructed by M. Delenil, of zine and protosulphate of mercury, 
which, being charged with pure water, exhibited for three months an action 
perfectly constant. 
The properties-of gas have formed the subject of several communications. 
M. Marcet called attention to some new experiments of M. Tyndall on their 
diathermal power. From these it results that simple gases absorb only 3 per cent. 
of the caloric emitted by the source of heat, while compound gases absorb 
much more considerable quantities of it ; olefiant gas, for instance, 81 per cent.; 
oxygen and nitrogen combined in protoxide of nitrogen, 60 per cent. Vapors 
absorb more than gases. M. Marcet likewise cited a memoir of M, Franckland 
on the influence of the rarefaction of air upon combustion, and showed that it 
may be considered as‘nothing. On the other hand, the intensity of light di- 
minishes rapidly with the density ; this loss of brightuess being, in England, 
0.05 to the inch of the pressure of mercury. M. L. Soret repeated before the 
society an experiment of M. Deville, intended to show a property of endos- 
