NATURE 
261 

THURSDAY, FEBRUARY 2, 1871 

SPECTROSCOPIC OBSERVATIONS OF THE 
AMERICAN ECLIPSE PARTY IN SPAIN 
| oaege complete and accurate account of the observa- 
tions of the American Eclipse party in Spain, under 
the charge of Prof. Winlock, which was given by Prof. 
Langley in NATURE two weeks ago, renders any further 
report of our spectroscopic work almost unnecessary ; and 
yet perhaps a somewhat fuller statement with reference to 
one or two points may not be wholly superfluous. 
Of the four spectroscopes employed, two were what 
might be called azalysing, and two integrating instru- 
ments. The analysing instruments are designed to study 
separately the spectrum of different portions of the 
prominences and corona, only a small part of the object 
being examined at a time; the integrating instruments, 
on the other hand, deal with the entire mass of light 
received from the whole luminous body, without distinc- 
tion of parts. In the first class of instruments, a distinct 
image of the object is thrown upon the slit by the object- 
glass of a large telescope, each point in the slit receiving 
light from only one point in the object. In the second 
class no image is thrown upon the slit, every point of 
which receives light from every point in the object ; and 
if a telescope is used at all (as was the case with Mr. 
Pye’s instrument), it must have a field of view large 
enough to include the whole object, and must have its 
eye-piece adjusted for distinct vision of a star—ze., in 
such a manner that the rays from a star shall be parallel 
when they leave the eye-piece. In this case the telescope 
increases the angular diameter and area of the object, 
and consequently the amount of light received, without 
otherwise at all changing the conditions. 
Prof. Winlock’s instrument and my own were of the 
former class. Prof. Winlock had a spectroscope of two 
prisms attached to an equatorial of 52in. aperture, and 
abcut 7ft. focus. My own instrument had (during the 
totality) the dispersive power of seven prisms, and was 
attached to an equatorial of 6; in. aperture, and 8ift. 
focal length. 
The instruments of Messrs. Abbay and Pye were of the 
integrating kind. Mr. Abbay’s had the collimator and 
observing telescope and two of the prisms which belonged 
to my old five-prism instrument. It was provided with 
the means of comparing the observed spectrum directly 
with the spectra of hydrogen, sodium, magnesium, and 
iron, 
Mr. Pye’s instrument was much smaller, but as its 
prism was made of the extra-dense flint, its dispersive 
power was very nearly the same as Mr. Abbay’s ; and the 
addition of a small telescope in front of the slit, magni 
fying about 2} times, and thus increasing its light about 
six-fold, made it, I think, fully the equal of the other in 
power and efficiency. 
Professor Langley has so well stated what we saw, that 
it is not necessary to repeat it; but I cannot refrain from 
putting on record that the sudden reversal into brightness 
and colour of the countless dark lines of the spectrum 
at the commencement of totality, and their gradual dying 
out, was the most exquisitely beautiful phenomenon 
VOL. III. 
| possibte to conceive, and it seems to me to have consider- 
| able theoretical importance. Secchi’s continuous spectrum 
at the sun’s limb is probably the same thing modified by 
atmospheric glare ; anywhere but in the clear sky of Italy 
so much modified indeed as to be wholly masked. 
I wish at this time to call special attention to the 
evidence which we obtain as to the extent of the self- 
luminous corona, or “leucosphere,” as it has been recently 
named,* by combining the indications of the two classes of 
instruments, 
By my direction Mr. Pye recorded the brightness of the 
lines which he saw during totality on an arbitrary scale from 
1o down, These are his numbers, C 8°5, D, 5°5, 1474 10, 
F 3. I suppose the actual amount of light of each 
kind would be roughly proportional to the squares of these 
numbers, for we seem instinctively to call one luminous 
object twice or thrice as bright as another when it would 
give the same light at twice or thrice the distance. 
If so, the numbers representing the relative amounts of 
light would stand C 72, D, 30, 1474 100, and F 9, neg- 
lecting fractions. 
Now, in the analysing spectroscope the case is very 
different, and it is difficult to make an accurate estimate ; 
but I think those who have been accustomed to observe 
both C and 1474, would admit that their ratio of bright- 
ness is something the same as that between a first and 
fifth magnitude star ; ze., Cis at least 25 times and per- 
haps 50 times as bright as 1474. Even during the totality, 
1474 can hardly be called conspicuous in an analysing in- 
strument, while C blazes like ared Sirius. It seems neces- 
sary,therefore, to assume that the area which emits the 1474 
light is to the area which gives C, roughly in the propor- 
tion of 100 X 25 (or 50) to 72—that is to say, the angular 
area of the self-luminous corona is from 35 to 70 times 
as great as that of the red stratum of hydrogen and promi- 
nences combined. I suppose these taken together would 
be about equivalent to a ring 15” high surrounding the 
sun, and this would make the self-luminous corona equiva- 
lent to another ring from 8’ to 16’ high. 
Of course I am aware that the numerical data of this 
calculation are very uncertain, and I have therefore 
neglected all considerations of shading and inequality of 
illumination. But the principle is, I think, correct, and it 
has this advantage. The presence of a light cloud or 
haze does not sensibly affect the result, because the calcu- 
lation is based solely on the ratios between lights of two 
different kinds in the two different instruments, and these 
ratios would not be seriously affected unless the cloud 
absorbed one kind of light more than the other. 
With the analysing spectroscope alone the case is en- 
tirely different ; a light cloud or haze vitiates everything. 
Thus some of the observers, favoured with a less clear 
sky than we at Xeres, saw the C and F lines even on the 
moon, undoubtedly by reflection from thin clouds. I 


saw myself the C line as far as 6’ or 7’ from the sun, far 
above any possible hydrogen atmosphere. 
Therefore, although Prof. Winlock and myself both saw 
the 1474 line to a distance of more than 16’ from the sun, I 
should not dare to lay much stress on that observation as 
showing the true limits of the self-luminous coronal matter. 
I base my belief that the limit of 15’ or 20’ is reached by it in 
* This name seems inadmissible, except as one of the sub-divisions of the 
chromosphere.—Ep. 
i 
