112 
Proceedings of the Royal Society 
ments of the zodiacal light. These are, of course, only approxi- 
mate, as the light has no definite boundary-lines, but gradually 
fades off at the borders. The measurements were made by the 
help of stars, as it was quite impossible to measure the light itself. 
On the 26th sunset was at about six o’clock, the light was very 
bright for 30° from the horizon, having at the horizon a breadth 
of about 25°, and at an altitude of 30° a breadth of about 20°. At 
9 p.m. the light could be traced quite round to the eastern horizon, 
a phenomenon which I observed on several other occasions. At 
10 p.m. the light was scarcely, if at all, visible. That this sudden 
disappearance was not caused by any change in the atmospheric 
conditions was clearly shown by the undiminished brightness of 
the stars. On February 27th the breadth at the horizon was 30°, 
while at an altitude of 30° it was only about 20°. The centre of 
the band passed a little to the south of the Pleiades. I endeavoured, 
by means of the sextant, to measure the inclination of the band to 
the vertical. For this purpose I chose two bright stars near the 
centre of the band — one at a considerable altitude, the other close 
to the horizon ; I then measured the angle between these and the 
angle between the upper one and the horizon ; these angles were 
respectively 55° 30' and 51°, giving the value of 31° 30' for the 
inclination of the centre of the band to the vertical. I had un- 
fortunately no access to star-charts, else I could have fixed the 
direction more accurately ; but even these rough observations con- 
firm the ordinary statement that the direction is slightly inclined 
to the ecliptic. 
The spectroscope used was one of Mr Ladd’s admirable small 
direct-vision spectroscopes, with five prisms and a single lens. 
Behind the slit is a small round hole, through which the light may 
be made to enter by opening the slit very wide, and which is very 
convenient for examining monochromatic light. With this in- 
strument it is easy to see a number of the Fraunhofer lines on a 
tolerably clear moonlight night. And such an instrument, though 
in some respects inferior to a simple large prism and slit, will, I 
believe, be found very suitable for work such as that described 
above, especially when it has to be carried on on board a ship 
much given to rolling. 
