Zodiacal Light. 167 



with the variations in the length of twilight, being an hour 

 and a half after sunset, at the vernal equinox, and two hours 

 and a quarter after sunset at the summer solstice. But were 

 these causes combined, the only or the chief reason why the 

 apparent elongation of the zodiacal light from the sun is 

 greater at one time than at another, then, since at the ver- 

 nal equinox the elevation above the horizon is at its maxi- 

 mum and the duration of twilight at its minimum, the ap- 

 parent elongation ought to be greatest of all ; whereas it is 

 then only 60°, while, from the 21st of November to the 18th 

 of December 1837, we found it increase from 90'' to 120°, and 

 this at a season of the year when the elevation above the 

 southern horizon is near its minimum, and the duration of twi- 

 light is longer than before. Nor is this an anomalous fact ; 

 the elongation has uniformly appeared greater in the west dur- 

 ing the months of December and January than during March 

 and April. Again, at the winter solstice, the elevation is much 

 greater in the morning than in the evening ; but the light 

 is far more conspicuous in the west than in the east. 



2. Direction. — The general direction of the zodiacal light 

 is, as its name imports, from the sun along the zodiac. 

 Cassini and Mairan thought that its axis lay nearly or quite 

 in the plain of the solar equator, making an angle with the 

 ecliptic of 7i° ; and, accordingly, that its nodes must be in 

 the part of the ecliptic which the earth traverses in June 

 and November, But Cassini himself remarked, that the 

 direction of the axis is not always the same. On several 

 occasions the vertex appeared to him to veer to the north- 

 ward of its previous direction, so that, while it would at one 

 time ]\x^i graze Alpha Arietis on its northern border, shortly 

 afterwards that star would be wholly within it. Before I 

 had met with these statements in Cassini, I had several 

 times remarked the same changes in the direction of the axis, 

 the vertex sometimes lying in the ecliptic itself. Nor, as I 

 think, will the observations warrant the conclusion that the 

 axis of this body cuts the sun, and, consequently, lies across 

 the ecliptic in the plain of a great circle. On the 19th of 

 January 1835, the northern border was 8° south of Castor, 

 and the vertex directed to a point south of the Pleiades. 



