SCIENCE. 
107 
\ 
latter augments and the line disappears, the ultra-violet 
flutings gradually die out altogether. 
It is philosophical to infer from these observations that 
not only are the line and flutings in question produced by 
carbon, but that the blue line (4266), since it is visible at 
the highest temperature, corresponds to the most simple 
molecular groupings we have reached in the experiments, 
and the flutings to others more complex. 
The result to which attention is most to be directed in 
this place is that touching the two sets of flutings, and 
should future research justify the double conclusion (1) 
that these flutings are truly due to carbon, a result I ac- 
cept, though it is denied by Angstrom and Thalen ; and (2) 
that the different flutings really represent the vibrations of 
different molecular groupings ; a great step, and one in the 
direction of simplification, will have been gained. 
Indeed it is much to be hoped that this ground will be at 
once worked over again by men of science who are both 
honest and competent : that the truth is sure to gain by 
such work is a truism. 
I have so often taken occasion to refer with admiration 
to the work of Angstrom and Thal&n that I shall not be 
misunderstood when I say that their conclusions, to which 
such prominence is given, and on which such great stress 
is laid by Messrs. Liveing and Dewar, rest more upon the- 
ory and analogy than upon experiment. 
Their work, undertaken at a time when the existence of 
so-called “ double spectra” was not established upon the 
firm basis that it has now, and when there was no idea that 
the spectrum recorded for us the results of successive dis- 
sociations, gave, as I have previously taken occasion to 
state, the benefit of the doubt in favor of flutings being due 
to compounds, and it was thought less improbable that 
cyanogen or acetylene should have two spectra than that 
carbon or hydrogen should possess them. 
Indeed, later researches have thrown doubt upon the 
view that the fluted spectra of aluminium and magnesium 
are entirely due to the oxides of those metals instead of to 
the metals themselves — and this is the very basis of the 
analogy which Angstrom and Thalen employed. 
The importance of the observations to which I have re- 
ferred is all the greater because of the general conclusions 
touching other spectra which may be drawn from them. 
Thus from what I have shown it will be clear that if my 
view is correct, the conclusions drawn 1 by Messrs. Liveing 
and Dewar from the assumed hydrogen-carbon bands 
touching both the spectrum of magnesium and the spectra 
of comets, are entirely invalid. These conclusions are best 
given in their own words : — 
“The similarity in the character of the magnesium-hy- 
drogen spectrum, which we have described, to the green 
bands of the hydrocarbons is very striking. We have sim- 
ilar bright maxima of light, succeeded by long drawn-out 
series of fine lines, decreasing in intensity towards the more 
refrangible side. This peculiarity, common to both, im- 
pels the belief that it is a consequence of a similarity of 
constitution in the two cases, and that magnesium forms 
with hydrogen a compound analogous to acetylene. In this 
connection the very simple relation (2 : 1) between the 
atomic weights of magnesium and carbon is worthy of note, 
as well as the power which magnesium has, in common 
with carbon as it now appears, of combining directly with 
nitrogen. We may with some reason expect to find a mag- 
nesium-nitrogen spectrum. . . . 
“ The interest attaching to the question of the constitu- 
tion of comets, especially since the discovery by Huggins 
that the spectra of various comets are all identical with the 
hydrocarbon spectrum, naturally leads to some speculation 
in connection with conclusions to which our experiments 
point. Provided we admit that materials of the comet con- 
tain ready-formed hydrocarbons, and that oxidation may 
take place, then the acetylene spectrum might be produced 
at comparatively low temperatures without any trace of the 
cyanogen spectrum or of metallic lines. If, on the other 
hand, we assume only the presence of uncombined carbon 
and hydrogen, we know that the acetylene spectrum can 
only be produced at a very high temperature, and if nitro- 
gen were also present that we should have the cyanogen 
spectrum as well. Either, then, the first supposition is the 
1 Paper read February 12, 1880. 
true one, not disproving the presence of nitrogen, or else 
the atmosphere which the comet meets is hydrogen only, 
and contains no nitrogen.” 
The importance of the question here treated of comes out 
very well from these two extracts. We find the same spec- 
tral phenomenon at once called into court, and very prop- 
erly called in, both to suggest the existence of chemical 
substances of which the chemist has never dreamt, and to 
explain the chemical nature of a large group of celestial 
bodies. 1 
There is little doubt that when a complete consensus of 
opinion is arrived at among the workers, other suggestions 
more far reaching still will be derived from the prosecution 
of these inquiries. For the present, however, the chief 
point to bear in point is that both in line-spectra and in 
fluted spectra we have indications which I think favor the 
view that in each case the origin is compound rather than 
simple. — Nature. J. Norman Lockyer. 
Oban, July 20. 
PHYSICAL NOTES. 
From the above article we see that as far back 
as 1878, Mr. Lockyer communicated to the Royal 
Society a paper in which the conclusion was drawn 
that vapor of carbon was present in the solar at- 
mosphere. This inference was founded upon experi- 
ments similar to those of Attfield and Watts, who 
showed that flutings are always present in different 
compounds of carbon exposed to the action of heat and 
electricity. This observation of Lockyer has been called 
in question by Liveling and Dewar, as they have found 
it an almost impossible problem to eliminate hydrogen 
from masses of carbon. This latter view has been long 
held by Edison, who, in a great number of experiments, 
some of which were participated in by Prof. Young, has 
found at the enormous heat developed by igniting a fine 
carbon thread ro%u of an inch diameter, of high resist- 
ance, in air vacuum, until a light of 80 candles is reached, 
that only a carbon spectrum is given, until just a few 
seconds before the rupture of the loop, when a sharply 
defined hydroge 7 i spectrum is observed. On the other 
hand, in an observation of the purified spectrum of car- 
bon tetrachloride, Mr. Lockyer ( Nature , August 5th) 
found only carbon appeared at high temperatures. It is 
an excellent index of the spirit of unbiased investigation 
in the author of {Nature, December, 1878) The Hypo- 
thesis that the so-called Elements are Compound Bodies, 
and still later, of the Universal Hydrogen Hypothesis, to 
learn from Mr. Lock>er that, both in line and fluted 
spectra, he thinks we have indications which favor the 
view that in each case the origin is compound rather 
than simple. 
In a communication from William Huggins, F.R.S., re- 
ceived June 16th, 1880, and published in the American 
Journal of Science for August, are embodied some observa- 
tions on the nature of the spectrum of water, which may 
give rise to a question of priority. It appears that Dr. 
Huggins made a photograph of the flame of hydrogen burn- 
ing in air, December 27, 1879, but did not publish the fact. 
On June last, Messrs. Liveing and Dewar state, in a 
paper read before the Royal Society, that they have ob- 
tained a photograph of the ultra violet part of the spectrum 
of coal gas burning in oxygen, and in a note dated June 
8th, they add that they have reason to believe that this re- 
markable spectrum is not due to any carbon compound, 
but to water. Professor Stokes (whose well-known mono- 
graph in Phil. Trans., 1852, has furnished so much sugges- 
tive material for others to work upon in this very line), 
authorizes the statement that Dr. Huggins, in a let- 
1 With special reference to this last question, that of cometary spectra, 
one of acknowledged difficulty, I may perhaps be permitted to add here 
by way of note that the view I put forward some years ago touching the 
relation to this spectrum to that of the nebulae has been lately strength- 
ened by the observation that at a low temperature one of the brightest 
lines in the spectrum of iron is that coincident with the chief line in the 
nebula-spectrum. 
