Fan. 19, 1871] 
NATURE 
223 

of the flag-ship, Zord Warden, two drawings were 
made, one by Captain Brandreth, and the other by Dr. 
Macdonald, F.R.S., in which the variation is so strong 
that one would feel inclined to acquit the atmosphere of 
any participation in the matter, and to relegate the whole 
outer corona to subjectivity alone, did not Mr. Brothers’s 
admirable photographs show both phenomena, as I am 
told they do. Dr. Macdonald saw eight rays arranged 
with perfect symmetry ; Captain Brandreth saw only two 
elliptical hoops crossing each other at right angles. 
Captain Cochran, of the Caledonia, besides the ring, 
saw a complicated stellate figure, the rays of nearly equal 
length, while Mr. Dexter, at sea between Catania and 
Syracuse, saw, besides the ring, ov/y one ray of inordinate 
length. 
So much for the drawings. I think that if the records 
of former eclipses be now examined, especially Mr. Car- 
rington’s drawing of the eclipse of 1851, and compared 
with the others taken at the same time, additional evidence 
will be gathered in favour of the compound nature of the 
corona, which, on the evidence now before me, I consider 
the great teaching of the present eclipse. Our experience 
in Sicily seems to be similar to that of the Spanish ob- 
servers, for Mr. Perry writes that “some observed two 
curved rays,” while the rapid degradation of light occurred 
at one-fifth of a solar diameter, but, so far as I know, no 
one in Sicily was favoured with a view of the dark intervals 
which were observed in Spain. 
There is a strange and most interesting discordance 
between some. of the spectroscopic observations made in 
Sicily and Spain. At Agosta, where the totality was well 
visible for ten seconds, Mr. Burton detected a green line 
near E, with a tangential slit (distance from moon not 
stated). This line, which was also seen by the Italian 
observers, is doubtless the one recorded last year by the 
American astronomers, but in Spain Mr. Perry states 
that bright lines at C near D, 4 (or E) and F were 
observed 8’ away from the sun, At Syracuse Prof. 
Harkness, whose telescope was moved into the various 
positions by Captain Tupman, R.M.A., found the green 
line in all parts of the corona, so far as about 10’ from 
the sun, and at one point thought he detected two green 
lines, less refrangible than it ; but at several places he saw 
a complete hydrogen spectrum (including C), which he 
attributed to prominences, until he was informed by 
Captain Tupman that there was no prominence near the 
_ slit. More proofs of the terrestrial nature of this portion 
of the corona, I think, taken in connection with the fact 
that the dark moon gave identically the same spectrum. 
It would appear that there was so much atmospheric 
reflection in Spain, and here and there at Syracuse, that 
the true coronal spectrum with its line near E, the 
existence of which we must now accept as established 
beyond all question, was partially masked by the promi- 
nence spectrum with its usual well-knownlines. There is 
one passage in Mr, Perry’s interesting letter in which, if 
there be a misprint, as I suspect there is, an observation 
of great importance is recorded. It runs, “ Mr. Abbay, 
observing at Xeres with a spectroscope of 2 prisms of 45° 
belonging to Professor Young, saw the bright lines C, D, 
F ; and afterwards F and a line rather more bright than 
'F on the less refrangible side of B, C not noticed then.” 

Now, if 6 (not B) was intended here we have sub incan- 
descent hydrogen mixed with the green-line-giving-sub- 
stance, which may probablybe a new element with a 
vapour density less than hydrogen. 
So that roughly we might regard the chromosphere to be 
built up of the following layers, which are in the orders 
of vapour density in the case of known elements :— 
X’(newelement) . . Green coronal line 
Hydrogen § Sub-incandescent F 
( Incandescent C, F, near G, h 
X(newelement) ... .. NearD 
Magnesium § 6 and lines in blue 
(and violet 
Sodium. 
Barium . Several lines 
{ Several lines, in- 
I ZC. ) 2 
es cluding E 
The foregoing table excludes naturally the substance 
or substances which give bright lines in the solar spectrum, 
which are at times visible in the spectrum of the chromo- 
sphere. I have ventured to suggest that the substance 
which gives the line in the green is a new element, because 
invariably I have found that in solar storms the chromo- 
spheric layers are thrown up in the order of vapour 
density, and because all the heavier vapours are at or 
below the level of the photosphere itself. 
With regard to the question of polarisation, the parties 
in Sicily obtained evidence that the corona was radially 
polarised, though Professors Harkness and Eastman 
obtained a result which they explain differently. Mr. 
Ranyard, at Villamonda, and Mr. Peirce, jun., north of 
Catania, obtained identical results in favour of strong 
polarisation. Hence the solar corona, accepting these 
observations, not only radiates, but reflects solar light to 
us. A careful consideration of this fact, taken in connec- 
tion with the possible addition of a, so to speak, terrestrial 
corona to its light, may enable us to account for some of 
the observations, both polariscopic and spectroscopic, 
which do not at first appear to harmonise with those to 
which I have referred, notably those which give a pure 
continuous spectrum to the corona, and which state that 
its light is only slightly polarised. 
From what has preceded, then, we seem justified in 
suggesting as working hypotheses the following, which, 
however, more accurate information may alter, and which 
I offer as suggestions only, dzex entendu. 
1. The Solar Chromosphere extends some 5’ or 6’ from 
the sun (Watson and others), its last layers consisting of 
cool hydrogen (Mr. Abbay), and possibly a new element 
with a green line in its spectrum (Young, Burton, and 
others) ; which line, if it be identical with the auroral line 
as stated by Gould, may possibly be present in the higher 
regions of our own atmosphere. 
2. Outside this stratum the rays, &c., are for the most 
part due partly to our own atmosphere, partly to our eyes, 
for their shape varies ; they are seen by some at rest, by 
others in motion, and their spectrum is the same as that 
of the dark moon (Maclear). 
3. The white light of the chromosphere above the 
prominences, as seen in an eclipse, is due to its strong 
reflection of solar light, as shown by the polariscopic ob- 
servations (Ranyard, Peirce, jun., Ladd). 
