seen at Port&ea on the 29th of March 1848. 437 



rathei' wider than the large circle ; and iheir edges, instead of 

 being sharp and colourless, were slightly fringed on the outer 

 or convex side with red, and on the inner side with violet', 

 and these coloured margins, which were not visible near the 

 sun, became more decided the further they receded from him, 

 as if the bands were formed by the overlapping of innumerable 

 spectra having their red ends directed exactly from the sun ; 

 so that by their combination they would render each band 

 white where it had the same direction as the spectra, but more 

 and more coloured as it became more oblique to them ; its 

 central line however being white throughout its length. The 

 knot at the triple intersection was also colourless. 



A fourth luminous arc was formed immediately over the 

 sun, but not concentric with him, like the common halo. The 

 distance of its vertex from him was indeed the same as the 

 radius of an ordinary halo, or about 23 ; but this curve was 

 described with a longer radius, its centi'e being, as shown in 

 the figures, some degrees heloia the sun, if indeed it was not, 

 as I sometimes thought, a portion of an ellipse with its long 

 axis horizontal ; and I had afterwards an opportunity of con- 

 vincing myself that this eccentricity was not a mere perspective 

 deception. This arc was brighter and lasted longer than the 

 large horizontal circle, though it was not so much noticed by 

 most spectators from its inconvenient nearness to the sun. It 

 consisted of all the prismatic colours arranged with the red 

 underneath or next the sun. The colours were much more 

 decided than those of the common halo, though not quite so 

 positive as those of the primary rainbow, but much brighter 

 than either, especially at the summit of the arch, from whence 

 its intensity rapidly declined each way, so that it could not 

 be traced quite down to its intersections with the large circle. 

 There was no particular appearance where it crossed the 

 other arcs. 



Tlie only remaining feature of this first exhibition consisted 

 of two detached single spectra or mock-suns, placed at equal 

 distances to the right and left of the real sun, and on the white 

 horizontal circle ; but what I thought very strange, they were 

 not at its intersections with the coloured eccentric arch, but 

 rather nearer the sun. I thought at first that they might be 

 at the places where an ordinary halo would have crossed the 

 horizontal circle; but neither was this the case; for in the 

 afternoon, when such a halo appeared, these spectra were left 

 outside its circumference, and independent alike of the common 

 and the eccentric halo. These spectra had their red sides 

 next the sun, and their colours were as positive, though not 

 so bright, as those of the eccentric arch. 



