SPECTROSCOPIC OBSERVATIONS. 61 



photograph of the spectrum of the " flash," and this was 

 actually done by Mr. Shackleton in the eclipse of 1896, and in 

 1898 the attempt was repeated and crowned with great success 

 at nearly all the observing stations. 



It was no exaggerated sense of its importance which, in 

 recent eclipses, has placed photographs of the " flash " as 

 absolutely the most instructive observations that could be made. 

 For this narrow shell, passed over by the moon in about two 

 seconds, and therefore about 800 miles in depth, is the lowest 

 region of the sun's surroundings which we can study. If it be 

 true, as Sir Norman Lockyer argues, that our terrestrial 

 elements are broken up in the sun into bodies more truly 

 elementary still, here we shall find the evidence of it. On the 

 other hand, if there be a true "reversing layer," this is the 

 region where we ought to find it. 



Observations of the spectra which an eclipse can offer to us, 

 and above all photographs of those spectra for a photograph 

 can record the position and intensity of a myriad of lines as 

 easily as of one, and is not liable to make mistakes due to hurry 

 or nervous excitement are the most important which can be 

 made. And of .all the varied phenomena of an eclipse, the 

 greatest interest at the present time attaches to the " flash." 



The spectroscopes employed for this work are of many 

 different kinds. Usually when a spectroscope is employed for 

 astronomical work it is attached to a large telescope, taking 

 the place of the eyepiece. The typical form of photographic 

 spectroscope, or " spectrograph " as it is called for brevity, 

 consists of a slit, placed in the focus of the great telescope, a 

 collimating lens, to render the rays of light which have entered 

 through the slit parallel, a prism, train of prisms, or diffraction 

 grating, to disperse the light and produce the spectrum, and 

 a photographic lens and camera to secure a negative of the 

 spectrum when thus produced. 



Such an arrangement would be termed an " analysing spectro- 

 scope," for an actual image of the eclipse would be formed on 

 the slit plate of the instrument, and the light from such 

 portions of the prominences and corona as lay across the slit 

 opening, would be analysed by it, every point of light on the 

 slit giving its own spectrum. But it is equally possible to use 

 the spectroscope by itself, without attaching it to any large 

 telescope, but simply pointing it up to the sky towards the 

 eclipse. In this case no image is formed on the slit plate, but 

 the general light from the whole phenomenon of the eclipse 

 enters the slit together and gives rise to a single composite 

 >jn'ctrum. A spectroscope so used is known as an "integrating 

 -pt-ctroscope," and is of great value as a supplement to the 

 analyser. Thus, for instance, in the eclipse of 1869, Young, 

 observing with an analysing spectroscope, found the green line, 



