To return to the protuberances. During the total obscuration, I was much impressed by the extreme brilliancy of their spectral lines. The idea immediately occurred to me that they might be seen even when the sun was unobscured; unfortunately the weather, which became cloudy after the eclipse, did not allow me to try the experiment on that day. During the night, the method and the means presented themselves clearly to my mind. Rising the next morning at three, I prepared for these new observations. The sun rose quite clear; as soon as it had risen from the haze of the horizon, I began to examine it, placing the slit of the spectroscope, by means of the finder, upon the same place where, the day before, I had seen the protuberances.
The slit, being placed partly on the solar disc and partly outside, gave, of course, two spectra, that of the sun and that of the protuberances. The brilliancy of the solar spectrum was a great difficulty; I partially avoided it by hiding the yellow, the green, and the blue portions, which were the most brilliant. All my attention was directed to the line C, dark for the sun, bright for the protuberance, and which, coming at a rather faint part of the spectrum, was seen with comparative ease.
I had not examined the right hand or western part of the protuberant region long when I suddenly noticed a small bright red line, forming an exact prolongation of the dark line G of the sun. Moving the slit so as to sweep methodically the region which I was exploring, this line remained, but changed its length and its brilliancy in the different parts, showing a great inequality in the height and brightness of the various parts of the protuberance. This examination was resumed at three different times, and the bright line always appeared in the same circumstances. M. Redier, who assisted me with much interest in these experiments, saw it as well as I, and soon we could even predict its appearance by merely knowing what region we were examining. Soon after, I ascertained that the line F showed itself simultaneously with G.
In the afternoon, I returned to the region examined in the morning; the bright lines again showed themselves, but they indicated great changes in the distribution of the protuberant matter; the lines broke up sometimes into isolated fragments which would not unite with the principal one, notwithstanding the shifting of the slit. This suggested the existence of scattered clouds formed during the forenoon. In the region of the great (or left hand) protuberance, I found some bright lines, but their length and arrangement showed that great changes had also occurred here.
These first observations already showed that the coincidence of the lines G and F was real, and that hydrogen was certainly the most important element in these circumsolar masses. They also established the rapidity of the changes which these bodies undergo, which cannot be perceived during the short duration of an eclipse.
The following days, I availed myself of all the opportunities allowed by the weather to apply and perfect the new method, at least as far as was permitted by the character of the instruments, which had not been constructed to suit this new idea.
Observing very attentively the lines of the protuberances, I have sometimes noticed that they penetrated into the dark lines of the solar spectrum, showing thus that the protuberance extends over part of the sun's disc. This result was naturally to be expected; but the interposition of the moon has always made its proof impossible during eclipses.
I will also detail here an observation made on the 4th of September at a favorable time, which shows how rapidly the protuberances change their form and position.
At 9h. 50m., the examination of the sun showed a mass of protuberant matter in the lower part of the disc. To determine its shape, I used a method which may be called chronometric, since time is employed in it as the standard of measure.
In this method, the telescope is placed in a fixed position, so chosen that by the diurnal movement of the sun all parts of the region to be explored shall come in turn into the field of the spectroscope; and at determinate times the length and situation of the spectral lines successively produced are noted.