320 
PEOFESSOli H. B. DIXON ON THE MOVEMENTS 
By tlie kindness of Dr. Schuster 1 was enahled to nse his apparatus to photogTaph 
the spectrum of the flame of cyanogen and oxygen as it exploded in a tube. With 
a sijigle prism, a continuous spectrum was obtained showing a few bright lines. In 
order to get a detailed spectrum, the gases were fired in a gun-metal tube end-on to 
the spectroscope, the end of the tube next the spectroscope being open just before ' 
firing. By this means sufficient light was secured to employ considerable dispersion. 
A cadmium spark spectrum was taken on the same plate for comjjaiison. No reversal 
of any line Avas shown on the })]iotograpli as in Livetng and Dewar’s experiment 
where the explosion tul)e Avas closed by a quartz plate. 
(2.) Photogrcvpluc Records of the Moving Flame. 
The discoAmi'A" of tlie detonation-AAuiA'e in mrseous mixtures bv Berthelot and 
ViEiLLE AA-as folloAA^ed shortly by the researches of Mallard and Le Chatelier on 
the initial phases of the explosion. The method they found most suitable for tracing 
the progress of the flame aatis a photographic one; the moAmment of the flame along 
a horizontal tulje being recorded on a sensitised paper moving Amrtically. Failing to 
obtain any photographic image of the flame Avith mixtures such as carbonic oxide 
with oxygen, Mallard and Le Chatelier employed caihon disulphide Avith oxygen 
or nitric oxide, these mixtures Aueldino- hiffiiD actinic flames. 
When the gases Avere igjiited by a flame at the open end of a long tube, the flame 
was propagated along the tube for some distance Avith a uniform sIoav A'elocity which 
Mallard and Le Chatelier regard as the true rate of propagation liy conduction. 
In the case of mixtures of carlion disulphide Avith nitiic oxide, this period of uniform 
moA-ement is succeeded by oscillations of the flame, AAdiich sometimes liecome of larger 
and larger amplitude and then die doAvn, and sometimes glA'e rise to the detonation- 
AAmve. When caiLon disulphide is mixed Avitii oxygen the preliminarv period of uniform 
moAmment is shorter and is immediateh" succeeded 1)A" the detonation-AvaA’e. Tliese 
xJ K' 
tAvo mixtures appear to be typical of other gase(Aus mixtures, cai'hon disulphide AA'ith 
oxygen resemldlng oxygen mixtures generally, carbon disulphide AA’ith nitric oxide 
resembling air mixtures g’enerallAU 
Mallard and Le Cha'J'EEIER draAv attention to the fact that in the explosions 
starting at the open end of the tube the deAmlopment of the detonation-AA’aA’e is not 
progressiAm, hut ahvays instantaneous. The detonatlon-AvaA-e is characterised not 
only by its great Amlocity of moAmment, hut liy its intense luminosity and the A-ery 
high pressures instantaneously set up in it. 
On the other hand, Mallard and Le Chatelier found that, Avhen the mixture of 
caiLon disulphide and nitric oxide AAms fired near the closed end of the tube the 
moAmment of the flame AA^as uniformly accelerated, until the detonation-AvaAm aa'us set 
up. Their apparatus, in AA'liich the photogi'aphic paper could not be moA'ed faster 
than 1 metre per second, failed to analyse the more rapid moA'ements of the flame. 
