120 GAS. 
other parts of our work,) reckoned upwards and down- 
wards from note C, to which the experiment pipe in 
GAS 
Gas Lights. water as will give it the consistence of treacle; make 
“== both the surfaces where the lid fits the retort very clean, 
spread the luting thinly ever the turned part, then se- 
cure the lid in its place by the wedge, Fig. C Plate | 
CCLXIV. Ifthis is not strictly attended to, the re- 
tort will lose the gas, and the smell will be very of- 
fensive, and injure the health of the operator. 
2d, The bridge of bricks, 6 4, Fig. 2. which sepa- 
“rates the fire place from the retort, must never exceed 
-a bright red heat. If they are raised to a white heat, 
-the gas will be injured, and the retort be soon destroy- 
ed. This may be regulated by a damper in the chim~ 
ney, or by the register door to the ash pit. See Fig. 7. 
3d, The gazometer should be examined at Jeast once 
a week, which is done.as follows: shut the stop cock y, 
Fig. 7. and likewise the retort from the gazometer, no 
operation going on. Mark the suspended vessel at the 
surface of the water when it is nearly full of gas. If 
the mark sinks below the surface, there is some open- 
ing where gas escapes. To find out this place, walk 
slowly round the vessel, and if the leak is not very 
small the gas may be smelled. Apply a candle to the 
place, and the issuing gas will be inflamed. Mark the 
place, and blow it out. In the same way search the 
vessel all round. There, however, may be a small leak, 
and yet it will not inflame. About the suspected part 
apply with a brush a little white lead paint. ‘The place 
where the gas escapes will become yellow, and ulti- 
mately black, from the sulphur in the gas. The place 
being discovered, take a small piece of linen, dip it in 
a little melted pitch and bees wax, and apply it to the 
part while hot, and keep pressing it on till it is cold. 
4th, Keep the water in the outer vessel of the gazo- 
meter at its proper height, in order that the gas may 
have to rise through the same column of water. 
5th, In the place where the lights are, appoint one 
person only to superintend their management. Be 
careful to shut the cocks when the lights are not want- 
ed, and do not suffer them to be opened till they are to 
be lighted, and then hold a lighted paper over the aper- 
ture while the cockiis turned. Do not nse acandle for 
this purpose, lest it drop.on the burner. (c. s.) 
GASES, Sounps propucep sy. In our article 
Acoustics, vol. i. p. 119, we have mentioned the ex- 
periments by Dr Chladni, on the sounds of different 
degrees of acuteness produced by the same organ-pipe, 
when blown with different gases, in appropriate re- 
ceivers ; and in page 118 we stated that the num- 
ber of vibrations which the same column of gases of 
different specific gravities should make in a given time, 
are inversely proportional to the square roots of their 
specific gravities. 
We propose, in the present article, to exhibit, in a 
tabular form, the results of the principal experiments 
that are recorded on this subject, for comparison with 
calculations on the above principles, and with other 
calculations from the velocities with which sound is 
propagated though different gases, considering’ the 
pitches of the sounds to be inversely proportional to 
the velocities of propagation, 
In the first column of our Table are mentioned the 
names of fifteen kinds of gases, on which Messrs F. Kirby 
and Arnold Merrick made repeated experiments, 
which are fully detailed, and their apparatus described 
in Nicholson’s Philosophical Journal, vol. xxxiii. p- 
171 ; and in the second and third columns are set down 
the mean results of these several experiments, as they 
have been calculated by Mr John Farey, in the Philo- 
sophical Magazine, vol. xlv. p. 28: The intervals in 
colums 2. being stated in his notation, (as usual in 
atmospheric air is supposed to be accurately adjust- 
ed. Column 8. shews the nearest notes on the Rey. 
Henry Liston’s Eusarmonre Organ, (see that article,) 
followed by the differences, whether more acute +, or 
ave —, expressed in Schismas; small and capital 
ftalic letters marking the octaves, above and ee 
Ce. 
In column 4, the ific gravities of the have 
been taken from ee it ‘chose sntutioned # our 
article Cuemistry; and column 5, (like col. 2.) shews 
the calculated intervals above and below C ; wherein 
it will be observed, that ether vapour, and sulphuretted 
hydrogen, appear to be graver notes than C, and ole- 
fiant gas more acute than it, contrary to the results of 
experiments thereon, in col. 2. 
In column 7, are contained the velocities with which 
sound is propagated, extracted from Acoustics, Vol. I. 
p. 122 ; in col. 8. are the intervals ; and in columns 6. 
and 9. the several notes and differences, as already de- 
scribed. 
By sub ing the intervals in columns 2, 5, or 8, 
(with due attention to the signs,) the relation or inter- 
val of any two gases may be found, and the name of 
such interval may often be obtained from our 30th 
Plate, in Vol. II. Thus, in the experiments.in col. 2, 
hydrogen appears to yield a sound higher than azote 
by 610 54-12 f+-53 m, or only 1.49 = less than an oc- 
tave. Again, the interval of azote and oxygen, in col. 5. 
is 55.2995 +f +5 m, or S—1.71 5; and een the 
sounds of oxygen and olefiant gases, is 55.702 4-f + 5m, 
or S—1.32, &e. 
The Table here presented will at least serve to shew, 
that much remains to be done, to reconcile the facts 
and the principles that have been advanced by different 
writers on the subject. Careful and numerous repeti-« 
tions of these experiments, with gases seottalyarepers 
and in well contrived apparatus, conducted as Mr Fa- 
rey has recommended, with reference by means of the 
beats, to fixed notes, carefully tuned on Liston’s organ ; 
not trusting to unisons for the comparisons of the 
sounds in any case, but resorting to the thirds and 
fifths by way of checks: Experiments so conducted, 
might, perhaps, so adjust these several intervals, that 
they may prove of use, in giving greater precision and 
consistency to the specific gravities, velocities of pr 
gated sounds, and perhaps to the weights of conan, 
of the several gases ; if it be true, as has been conjec« 
tured (and, as seems nearly true of most of the gases 
in our Table, and perhaps of others,) that, with the 
exception of oxygen and olefiant gases, the weights of 
atoms, nearly as stated by Dr Thomson in his Annals 
of Philosophy, are exactly double (or octave) of the 
specific gravities, respectively, (to oxygen, 1.); but ni~ 
trous gas seems here to form a remarkable exception, as 
Dr Chladni found it to present, on another point, in 
his experiments, as mentioned in our first volume, 
p- 119.. It may not be amiss to add, that the specific 
gravity of nitrous gas being 1.094, its note will be 
C’p—3.63 £=39.63 >4-f-4 8m, below C, according to 
the principle of calculation used in our Table ; whereas 
in Messrs Kirby and Merrick’s first set of experiments 
(see Phil. Mag. vol. xxxvii. p. 4.), this gas was ob« 
served to sound 52.955+f45m above C, It must 
however be observed, that the results of the first and 
second sets of experiments by these gentlemen,. are 
most of them so greatly different, as to shew strongly 
the necessity of the repetitions thereof that we haye re- 
commended above. 
