TOTAL SOLAE ECLIPSE OE JULY 18, 1860. 
.309 
the bright line was apparent, but it disappeared when the crescent was lifted up, and a 
sheet of white paper was interposed between it and the dark ground of the photograph. 
These phenomena occurred when the photograph was examined with the naked eye, 
with the aid of spectacles, or, from a short distance, with a sharply defining telescope by 
Ross. Viewed in either of these ways, the brightening was found to begin immediately 
beyond the edge of the white paper as it was introduced more or less under the crescent. 
For the purpose of illustrating this paper on the occasion of its being read before the 
Society, I prepared a representation of one of the photographs of partial phase, 3 feet in 
diameter, in which, bearing in mind the well-known fact that there is on the solar disk 
a gradual diminution of the intensity of the light from the centre to the periphery, I 
carefully reduced the brightness of the solar crescent in due gradation towards the con- 
vex boundary. In the first instance the background was not painted in, and I expected 
that when it was completed a brightening would immediately occur. Such, however, 
was not the case. 
On calling Professor Stokes’s attention to this failure in producing the phenomenon 
of brightening by artistic means, he suggested that I should renew the attempt by using 
a real photograph of the sun and a dark disk for the moon *. On this plan I succeeded 
in making eclipse-pictures artificially, which showed the brightening very distinctly. 
From these experiments I am inclined to believe that Mr. Aiet’s explanation is the true 
one, although it is a curious subjective fact that the parts possessing superior illumina- 
tion exhibit to our perception an extremely bright line, bordering immediately on the 
dark limb of the moon, while the less bright parts towards the circumference present 
no such appearance, although they also are contrasted with the dark background. 
In order to study other points connected with the photographs, I had made, on glass, 
some enlarged copies, in which the moon’s disk was increased in some cases to 9 inches, 
in others to 13 inches in diameter. It was found that measurements could be made on 
these with considerable accuracy, by means of a graduated beam compass reading to 
thousandths of an inch, and I had proceeded to some extent in this way, when it 
occurred to me to have an instrument constructed expressly for measuring the original 
negatives. The study of the enlarged copies led, however, to a method of producing 
charts of the prominences with complete fidelity ; and the plan will, I think, hereafter 
prove applicable to the production not only of astronomical, but also of other graphic 
representations derived from photographs. In order to carry out this method, a table 
had to be constructed with a square hole cut in it somewhat smaller than the glass 
positive to be worked upon ; a recess surrounding the hole was made in the top of the 
table, just the size of the glass, and of a depth corresponding to the average thickness of 
the plates. Four spring clips served to hold the glass firmly in its seat. Parallel with 
* Eor the lunar disk I employed photographic paper darkened to the same tint as the background of the 
solar photograph. These disks were in some cases neatly inserted in a circular hole in the solar picture, 
and in other cases pasted on it. In eitlier case the surface was polished and made uniform by passing the 
picture through a rolling-press. 
