( I2 9 )

and ttß y, the fecond Prism, which Refracts back again the Rays to Q_, whcrcthe long Image PT is contracted into a Round one. i soppofe the Planea.y Parallel toBC, and ß y to AC, that the Rays may be equallv Refractedcontrary ways in both Prifms. 1 he Prilms also must be placed very near toone another ; for It their Diltance be sogreat, that Colours begin to sppearin the Light, before its Incidcnce on the lecond Prisin, thole Coiours will notbe destroyed by the contrary Refractions of t hat Priliii. And if a Lens be placedin the hole F, orimmediarely aster the Prilfns, so that its Fccrs be at the I-mage Q_- or P T, the Perimeter ut the I mage Q,and the straight sides of theImage P T, will become nuich better defined than othcnvilc. So that, what- never was the caule ot that Length, ’twas not any Contingent Irregularity.

I then procecdcd to examine more critically, what might be cflecied bythe difterci.ee of the Incidcnce of Rays coming from divers parts of the Sun,and to that end, mealtired the scveral Lines and Angles, belonging to the I-mage, Its diitance from the Hole orPrilin was 22 foot, its utmoit Lengthi 3 4 inch.es ; it’s Breadth 2 f ; the Diameter of the Hole qp of an indi; theAngle which the Rays, tending towards the middle of the Image, made withthole Lines, in which they would have proceeded without Refraction, was44 deg. y6 min. And the Vertical Angle of the Prilm, 6 ] deg. jimin.Also the Refractions on both sides the Prilm, that is of the Incident and E-mergent Rays, werc, as near as I cotild make them, Equal, and conlequenc-ly about y_j. deg. 4 min. And the Rays feli perpendicularly upon the Wall.Now lubducting the Diameter of the Hole from the Length and Breadth ofthe Image, there remains 1 3 inches in the Length, and z $ the Breadth,comprehended by thole Rays, which paised through the Center of the liiidHole, and conlequently the Angle ot the Hole, which that Breadth lübtend-ed, was about 31 min. anlwcrable to the Sun's Diameter; but the Anglewhich its Length sobtended, was more than y such Diameters, namely z deg.49 min.

Having made thele Oblervations, I sirst computed from them the RefractivePower ot that Glals, and found it mealurcd by the Ratio of the Sines, xo to3 1; And theil by that Ratio, I. computed the Refractions of two Rays flowingtrom opposite parts of the Sun’s Discus , so as to dicter 31 min. in their Obli-quity ot Incidcnce, and found, that the Emergent Rays fiiould have compre-hended an Angle of about 3 1 min. as they did before they were Incident.

But becaule this Computation was founded on the Hypothclis of the Pro-portionality ot the Sines of Incidcnce and Refraction, which thougii by myown Experiencc Icould not imagine to be so erroneous, as to make that An-gle but 31 min. which in reality was 2 deg. 49 min. yet my Curiosity cuuledme again to takemy Prilm. And having placed it at my Window, as before,

I oblerved, that by turning it a little about its Axis to and fro, lo as to va-ry its obliquity to the Light, more than an Angle of 4 or y degrees, the Co-lours were not therehy lensibly Tranflated from their place 011 the Wall, andconlequently by that Variation of Incidcnce, the quantity of Refraction wasnot lensibly vavied. By this Experiment therefore, as well as by the formerVol. I. S Com-