254 
Now from very careful examination, he considered that 
neither the A line nor the B line was due to aqueous vapour, but he 
thought they might be due to carbonic acid, From what we have 
observed, it seems highly probable that they are not due to car- 
bonie acid because, if we had a compound of carbon and oxygen, 
we should not expect to see such a spectrum at all. I think that 
from the context we shall see a great probability as to what they 
may be due. Other observers observed the spectrum of the 
atmospheric lines, but perhaps the most recent and trust- 
worthy observer is Prof. Piazzi Smyth. He published in the 
‘* Edinburgh Astronomical Observations,” vol. xiv., a memoir 
regarding these black lines. Regarding B he says that they are 
merely dry air lines, and that the A line is a solar line. I 
should like to quote his own words with regard to this particular 
Aline. He says, ‘The remarkable clearness, plainness, and 
strength of the lines composing A and its parallel bands in ahigh 
sun tend to claim the said A and its appendages a solar line, and, 
if so, it is a grander one by far than any other solar line in the 
whole of the solar spectrum. If, on the other hand, some 
observers will claim A as only a gas effect, Angstrom has said that 
ah 
1 
Fic. 18.—Top spectrum the A line ; bottom spectrum the benzine radical 
band. 
A grows or reinforces at sunset. I may say that I have seen it 
this summer (1877) in Edinburgh almost at the point of sunset, 
and ascertained with a more powerful prism, that, thcugh heavily 
black as it was, it hardly was in reality grown at all.” He says 
that the time was too short for him to measure absolutely the 
number of lines, but his impression was that the number of lines 
composing the A band was not diminished at all, and that the 
A line presents the same appearance at sunset as it did in a high 
sun, 
Now, as I said at the beginning of my lecture, we must take 
into consideration the general absorption which is always more or 
less visible in the spectrum, and at sunset we always have greater 
general absorption than we have of course when the sun is at mid- 
day, because the stratum of air between us and the sun is greater 
in the one case than it is inthe other, Smyth, I believe, sup- 
posed that the A line was due to hydro-carbon, and I should like 
to show you that, probably, his surmise 1s correct. [I subsequently 
found that this idea did not originate with the Professor.] An 
examination will show you how very remarkably like the A group | 
NATURE 
| Fan, 12, 1882 
of lines is, to this B grcup ef lines. They both have a band—a 
commen band » hich, with a very high magnifying power, is 
splits up into fine lines. Frem analogy, one would suppose, 
then, that the A and B lires are due to the same substance. If 
one looks at the A line with a small dispersion, one sees it very 
much of the same form as the benzine radical band that you saw 
(Fig. 18). Itappears to have a dark distinct band, and then to shade 
off for some considerable distance. Now the remarkable thing 
about this is that if I transferred this A line such as we see here 
to about a thousand wave lengths down the spectrum, we should 
get the characteristic radical band of benzine, In our comparison 
of the solar spectrum with that of benzine we have feund unmis- 
takable proots of the existence of benzine in the solar atmosphere. 
This being the case, the big A line and the benzine, or the benzine 
series may have some connection one! with the other; and it 
seems to me from analogy that sooner or later the A line will 
be fcund to be due to a hydro-carbon cf some kind. You recol- 
lect in the photograph 1 showed you last time, that as the mole- 
cule got lighter the radical band went up towards the blue. Now 
the benzine has essentially a comparatively speaking heavy mole- 
cule, It is composed of 6 of hydrogen and 6 of carbon, and 
therefore if we can examine a hydro-carbon, which has the same 
proportions of hydrogen to carbon, but of a different molecular 
grouping, I think it extremely probable that eventually we shall 
find that the A line belongs to that particular group of the hydro- 
carbons in which there are equal atoms of hydrogen and carbon. 
| As to the B line the question is, canthe B line, which has a 
similar look to the A line be due to a hydro-carbon. That re- 
quires great consideration. Every observer, 1 believe, has put 
B down as an atmospheric line, end therefore if such be the 
| case we shall have to lock for a hydro-carbon high up in our 
atmosphere, 
Now, to come back, what would be the effect of aqueous 
vapour? First of all, we should have a general absorption in 
the ultra red part of the spectrum, and then also a special ab- 
sorption. The annexed diayram (Fig. 19) is from a photograph 
of the solar spectrum taken through various thicknesses of water, 
and you will see what kind of effect the various thicknesses have 
upon absorption 
This bottom figure is the spectrum through three feet of 
water. The next part of the spectrum is photographed through 
one foot of water; and the next part of the spectrum is photo- 
graphed through six inches of water. You will see what a re- 
markably little absorption there is of the visible spectrum due to 
a certain amount of water. Now, can water have the same effect 
as aqueous vapour? 1 think it can. If M. Janssen had con- 
densed the steam in the 330 feet tube, and only had a thickness 
of somewhere about three inches of water to look through, then 
looking through three inches of water he would have seen no 
absorption whatever. In other words, where you have a liquid 
converted into a vapour a given weight of it seems to absorb 
a great deal more as vapour “than it does as liquid. Sometimes 
we see this even in the visible part of the spectrum. 
When you havea cold north-east wind, you have an atmo- 
sphere very free from moisture, and you are ‘able to go down in 
the infra red regions very far; but when you have a change of 
CB Am ee. Oe 
Fic. 1u.—Absorption spectra of different thicknesses of water, 
weather and a south-west wind, for instance, you have a remark- 
able absorption setting in, and -you have the limits of the solar 
spectrum ending somewhere about 7, and you may try your 
utmost and yet on one of these days you will not be able to go 
one jot farther than that particular line. In other words, the 
general absorption due to atmospheric water in some form or 
another comes in, and not special absorption. 
In the diagram, Fig. 7 (see p. 188), we have two bands, o 
and y, at the extreme right of the ultra red, and only on two 
days in two years have T ever been able to obtain a photograph 
of those particular bands. They were obtained in both cases in 
March, and it was when the air was particularly free from 
aqueous vapours, and it was very cold. To sum up, where you 
| a genera 
have a west wind, or a south-west wind and high temperature, 
l absorption takes place in the ultra part of the spec- 
trum. It is only when you have very cold air, with a north-east 
wind or dry wind, that you are able to get even as far down as 
we have got. 
The next point to which I wish to call your attention is, Is it 
probable that in the ultra red part of the spectrum we should 
ever be able to obtain evidences of metallic lines, or lines which 
are due to metallic vapours? I think that it is highly improbable 
that we shall. The few metals that we have tried have given almost 
a negative result except in one or two cases. Thus in looking 
through the list of lines due to the metallic elements, there are 
| very few that go even as far as 700 in the scale of wave-lengths. 
