THE SOLAR ECLIPSE OF 1898. 
253 
between the outer streamers and tbe bright part near the limb is very 
great* while the outer region is that most difficult of photographic 
subjects* a faint object on an illuminated ground. As we require a 
large image it is necessary to use either a lens of long focal length ora 
secondary lens to magnify the image. 
The British observers have adopted a standard form of instrument* 
it being of advantage for purposes of comparison that the same instru¬ 
ment should b© used at different eclipses. This instrument is a double 
one* two cameras being mounted side by side in the same tube. One 
camera is provided with a lens of four inches aperture and about four 
feet focal length* thus giving an image of the sun about half an inch 
diameter. The other has a lens of nearly the same size* but a negative 
enlarging lens is added which magnifies the image so that the sun is 
H inches diameter on the plate. The plates are carried in the same 
holder* so that the exposure* which is made by a shutter* is simultaneous 
on both plates. Experience has proved that the larger image gives 
decidedly better results. Attempts have been made to obtain photo¬ 
graphs on a much larger scale than this. In 1893 Professor Schaeberle, 
from the Lick Observatory* wbo stationed himself in Chili* where the 
atmosphere was particularly clear* got some very fine photographs with 
a lens of 40 feet focus* thus giving an image of the sun about 4 inches 
diameter. The aperture of his lens was small* and ho consequently 
only got the brighter or inner corona on the plate. Last year other 
observers had arranged to try to get large scale photographs* but the 
unfortunate weather conditions frustrated their efforts. In future 
eclipses this will be an important feature. 
Direct photographs of the corona are valuable* not only as a record of 
the form of the corona at the time* but also in determining whether 
the corona shows any change during the actual duration of an eclipse. To 
this end we endeavour to place observers as widely separated as possible 
on the eclipse track with similar instruments* thus in the most favour¬ 
able circumstances getting an interval of three or four hours between 
the times of observation at the two stations. 
In 1893 photographs of this sort were taken in Brazil and Africa at 
an interval of about hours and showed no perceptible change in 
form of the corona* but it does not* of course* follow that at another 
eclipse time the corona might not be in a less stable condition and might 
not show a marked change of form as photographed at two stations far 
apart. In the cases of the prominences we know that they often 
remain unchanged for many hours* while at other times a large pro¬ 
minence may form or disappear in the course of half an hour. 
For investigating the spectrum of the corona and thus determining 
its composition we make use of a spectroscope of the ordinary form* 
using a lens to throw an image of the eclipsed sun on the slit. The 
results obtained hitherto are far from complete. We know that the 
corona is partially composed of glowing gas and partially—though only 
to a small extent—of fine dust* reflecting the light of the sun, but what 
the gases are that compose it we do not know* save that they do not 
appear to be gases known on the earth. It is obvious* therefore* that 
there still remains here a large*field for investigation, 
