1868. | Physics. 111 
state, and the spectrum of this gas being yet unknown, the observ 
ations of M. Lielege have served to fill up a gap in the series of 
spectra produced by the gases in ignition. ‘The apparition and 
the disappearance of some of the luminous fixed lines is closely 
connected with the metallurgical operation. At the moment the 
decarbonization of the iron is nearly terminated, the spectral lines 
undergo essential modifications. The apparition of a group of 
lines and of an isolated line in the violet-blue portion of the spec- 
trum marks a particular reaction, during which the soft iron is 
being formed, and these lines disappear sooner than all the others ; 
their appearance and disappearance serve therefore to indicate the 
termination of the process. 
The same indefatigable observer has also ascertained that the 
spectrum of the colour of sea water is deprived of its red portion 
at small depths, and successively of the yellow and green, for the 
greater depths, until it appears of a violet-blue. In trying to 
ascertain whether the same was the case in glaciers, he has made 
some interesting experiments in an artificial grotto in the Grinden- 
wald glacier. This cavern is 100 métres deep, transparent in its 
walls, through which the solar light penetrated. The light was of 
a fine blue tint, the red being extremely weak, so that in the grotto 
human countenances assumed a cadaverous aspect. On looking 
towards the entry, at a certain distance in the cavern, it appeared 
to be lit up with a red light, doubtless the effect of contrast. The 
thickness of the superposed mass was not enough to show a greater 
effect than the almost complete absence of the red, and a great 
diminution of the yellow. The ice was said to be 15 metres thick, 
but was probably less; it was perfectly compact and limpid, but 
with a few air-bubbles. 
M. Felix Lucas concludes from theoretical considerations that. 
the luminous distance at which the electric spark is visible is 
greater than that of a permanent light, the apparent intensity of 
which would be 250,000 times that of the spark. The light actually 
employed to illuminate modern lighthouses gives a brilliancy equal 
to 125 carcel lamps. An electric spark possessing the illuminating 
power of the 200th part only of a carcel burner is superior as to 
its power of projecting light. Hence we can conceive the immense 
effect of a warning light composed of intermittent flashes of the 
electric spark proceeding from a strong Leyden battery. M. Lucas 
states that, in an experiment made in a laboratory two apparatuses 
were established, one voltaic battery equal to 125 carcel lamps, and 
another spark-battery equivalent to only the 1-2000th part of a 
carcel lamp. The photometer (such as is employed in the light- 
ae administration) showed a marked superiority in favour of the 
spark. 
