NATURE 



237 



THURSDAY, JANUARY 25, 1872 



THE SOLAR ECLIPSE 



ACCOUNT OF OBSERVATIONS MADE AT POODOCOTTAH 



THE spectral observations of recent total eclipses of 

 the sun had plainly demonstrated the existence of an 

 incandescent gaseous stratum or atmosphere, surmounting 

 the chromosphere or stratum of hydrogen which envelops 

 the body of the sun, but they had not sufficed to deter- 

 mine its true conformation and extent. This question, 

 therefore, constituted one of the principal problems re- 

 maining to be solved by observations of the eclipse of 

 the i2th of December, 1S71. 



The slit-spectroscope applied to large telescopes doubt- 

 less affords the best means of verifying the existence, in 

 the circumsolar regions, of this gaseous stratum, which may 

 be termed the superior chromosphere, and of determining 

 the materials of which it is composed ; but from the 

 shortness of the time available in an eclipse, the spectro- 

 scope can furnibh only partial and local results, insufficient, 

 therefore, to reveal the true structure, form, and dimen- 

 sions of this upper chromosphere. 



Preceding observations having shown that the light of 

 the solar corona is composed for the most part of a small 

 number of elementary rays differing considerably in re- 

 frangibiUty, it appeared to me that the form and dimen- 

 sions of the higher chromosphere might be much more 

 conveni;ntIy studied by means of a large prism fixed in 

 front of the object-glass of the telescope, whereby the 

 several chromatic images of the corona would be distinctly 

 formed in the focal plane. If the prism has but little dis- 

 persive power, and the eye-piece does not magnify too 

 much, all the chromatic images of the corona may in 

 this manner be observed simultaneously in the same field, 

 and their form and dimensions directly investigated. 



Towards the end of the year 1868, a small flint-glass 

 prism was made for me by Signor Merz, of Monaco, to 

 be fitted to the object-glass of the equatorial belonging to 

 the Observatory of Campodogho, for observations on the 

 spectra of the stars ; and this apparatus, in consequence 

 of the dispersion of the prism, and the goodness of 

 this prism and of the object-glass, was found to be ad- 

 mirably adapted for observing the eclipse in the manner 

 just described. 



The dispersion of the prism from the lines C to H of 

 Fraunhofer is about 32' ; the free aperture of the object- 

 glass is 4j* French inches ; the field of the telescope is 

 about i", with a magnifying power of 40. 



My conviction of the great advantages which would be 

 afforded by this instrument in the observation of the ap- 

 proaching eclipse, induced me to carry it to India for that 

 purpose ; and I was glad to learn that Mr. Lockyer, the 

 chief of the expedition, had in like manner resolved to 

 observe the corona by means of a spectroscope without a 

 slit, being persuaded that this would be the most con- 

 venient method of solving the questions relating to the 

 corona itself With this instrument, then, I prepared to 

 observe the eclipse, proposing to myself the following 

 problems : — 



VOL. V. 



1. To ascertain whether, just before the beginning, and 

 at the end of totality, the solar spectral lines are reversed 

 — a phenomenon observed by Prof. Young in the eclipse 

 of 1870. 



2. Amidst the several chromatic images of the promi- 

 nences, to observe especially whether the image given by 

 the yellow line D^ coincides with that of the lines of 

 hydrogen gas. 



3. To define the form and dimensions of the chromatic 

 images of the corona. 



The day before the eclipse, I delineated, by means of 

 the direct-vision spectroscope applied to the telescope, 

 the profile of the solar disc, in order to ascertain the state 

 of the chromosphere at the several parts of the limb, and 

 the protuberances existing there. But the picture did 

 not come out with sufficient exactness, in consequence of 

 the cloudy state of the sky, and the strong wind which 

 prevailed throughout the day. This picture, however, 

 clearly showed that both on the eastern and on the 

 western limb, at the point where contact would take place 

 between the lunar and solar discs in the total eclipse, the 

 chromosphere was in that abnormal condition which 

 is generally observed in the neighbourhood of solar 

 spots. 



The number of the prominences was, however, rather 

 small, and their dimensions moderate ; conditions which 

 appeared to me to be favourable for the examination of 

 the corona. 



From the 5th to the nth of December, the state of the 

 sky at Poodocottah was somewhat variable ; and gene- 

 rally, in the early hours of the day, great masses of mist 

 and cloud predominated in the east, leaving but little hope 

 in favour of our station for observing the eclipse. On 

 the morning of the 12th, indeed, the sky was almost 

 wholly covered with dense masses of mist and cloud, 

 completely obscuring the sun till 7h. 53m., at which time 

 the eclipse had already begun. Soon after this the sun 

 was again covered with thick clouds, but fortunately they 

 began to break a few mjnutes before totality, when the 

 bright disc of the sun was already sufficiently reduced, 

 and when consequently the time for observation was 

 rapidly approaching. 



To verify the phenomenon of the reversal of the 

 spectral lines at the extreme edge of the sun, I had 

 arranged the plane of dispersion at right angles to the 

 edge at the point of second contact. 



At thirty seconds before totality, the spectral image of 

 the luminous crescent was already sufficiently weakened 

 to allow of its observation by the naked eye without a dark 

 glass ; and it was then that the principal dark lines of the 

 solar spectrum came out distinct, and even more strongly 

 marked than before, and curved parallel to the bright edge 

 of the sun ; but a few seconds before totality these lines dis- 

 appeared completely, and the spectrum became continuous, 

 without however exhibiting, just before totahty, the re- 

 versal of the lines, although I was watching most intently 

 for this phenomenon. I would not, however, be under- 

 stood as denying altogether the reversal of the lines, for 

 it is not impossible that a thin film of mist, or the bright 

 atmospheric light at that time diffused over the spectrum 

 of the solar limb, may have concealed the bright lines. 



At the very instant of totality, the field of the telescope 

 exhibited a most astonishing spectacle. The chromo- 



