( l 7° )

The reason of this difference, I apprehend, was, that in the clearest Days irtthe Light of the White Skies, which dilutes and renders invisible thc faintestColours at the Ends of the Image, is a little diministied in a ciear Day, andIo gives leave to the Colours to appear to a great Length; the Sun’s Lightat the fame time becoming Brisker, and sb strengthning the Colours, and ma-king the faint ones at the two ends more conlpicuous. For I have obscrved,that in Days something Cloudy, whilst the Prism has stood unmoved at theWindow, the Image would grow a little longer or a little Ihorter, accor-dingly as the Sun was more or leis obseur’d by thin Clouds which pafled overit; the Image being {hortest when the Cloud was brightest and the Sun’sLight faintest. Whence ’tis easie to apprehend, that if the Light of theClouds could be quite taken away, so that the Sun might appear liirround-ed with Darkness, or if the Sun’s Light were much stronger than it is, theColours would still appear to a greatcr Length.

In all thete Observations thc Breadth of the Image was- just two inches.But observing that the Sides of the two Prisins I- ufed were not exactly plain,but a little Convex, (the Convexity being about so much as that of a doubleConvex-Glass of a sixtcen or eighteen Foot Telesoope) I took a third Prisin,whose fides were as much Concave as thoso of the other were Convex; andthis made the Breadth of the Image to be two inches and a third part ofati inch ; the Angles of this Prisin, and the Lengths of the Image madc bye&ch o£ thosc Angles, being thoso exprest in this Table.

Angles

58°-

6i i

Leng bs8 a I ,

9 , '10 -3 !

ßt this case you see, the Concave figure of the fides of the Prisin by irn-Iting the Rays Diverge a little, causos the Breadth of the Image to be greaterin proportion to its Length than it would be othervviso. And this I thoughtfit to give younotice of, that Mr. Lucus may examine, whether his Prisin havenot this fault. If a Prisin may be had with sides exactly plain, it may dowell to try the Experiment with that; but its bester, if the sides be about somuch Convex as thoso. of mine are, because the Image will thereby becomemuch heiter defined. For this Convexity of the sides does the fame effect, asj£ you fhould ufe a Prism with sides exactly plain, and between k and the.Hole in the Window-lhut, place an Object-glasi of an 18 Foot Telesoope, tomake the Round Image of the Sun appear distinctly defin’d on the Wall whenthe Prisin is taken away, and confequently the Long Image made by the Prisin,to be much more distinctly defined (especially at its streight sides) than k wouldbe otherwise.

One thing more I (hall add: That the utmost Length of the Image,firom the faintest Red at one End, to the faintest Bitte at the other, must bemeasored. For in my first Letter about Colours, where I fet down the Lengthw be fiye times the Breadth, I called that Length the utmost Length of.. ' . the.