144 Proceedings of Boyal Society of Edinburgh. [sess. i 
line, apparently in the yellow-green, but position doubtful, as only a 
rough method of comparison was available. 
March 29th, 7.30 to 8 p.m. — Short collimator, low power eye- 1 
piece. Night fine: no clouds. Continuous spectrum, brighter part ! 
beginning in the yellow. Bright line appeared to flash out at times, 
but very doubtful. 
April 2nd . — Clouds near horizon. Bright continuous spectrum, 
which ended sharply near D. A place of maximum brightness 
near centre. Flickering. 
April 5th . — Direct vision spectroscope. Clouds near horizon. 
Light faint, but spectrum fairly bright. There appeared to be two 
bright lines. At times the lines seemed quite distinct, at other 
times quite invisible. 
April 24 th, 7.30 p.m. — Short collimator, high power eye-piece. 
Light clouds and a slight haze near the horizon, through which the 
Pleiades were dimly visible. Z. L. not very bright, but extended 
through at least 60°. Spectrum pretty bright, and bright line 
appeared to flash up near the red end. Position determined by 
bringing end of faintly illuminated scale to coincide, and then com- 
paring solar lines with the scale on the morning of the 25th, every- 
thing having been left unchanged. 
The position thus determined corresponded to a wave-length of 
558. After this the spectroscope was fitted for photographic work, 
and plates were exposed on April 30th and May 1st; on the former 
for an hour, on the latter for an hour and a half, hut without 
any result. The following is the note for the latter of the two 
dates : — 
May 1st. — Z. L. very bright but very diffuse, rising well above 
Jupiter, near which it seemed to bifurcate. Slight haze about 
horizon, but stars visible through it down to the horizon. Spectrum 
as seen with eye-piece held behind camera very bright, and extended 
to the red. With direct vision spectroscope very bright at first, but 
soon getting much fainter. The line at times seemed to flash out. 
At other times the spectrum seemed to be banded. 
Since then I have never seen the least trace of a line, though on 
a number of occasions I have had particularly good opportunities of 
examining the spectrum, especially in January 1885, when I spent 
the first twelve days of the year on the top of Dodabetta, 8642 feet 
