1Q8 PHILOSOPHICAL TRANSACTIONS. [ANNO 17 Q7* 



at the edge, and even pretty distinct at the end ; they were also much narrower 

 than those images, but like them they inclined much to the violet, and were broadest 

 in the red, becoming narrower by degrees, and narrowest of all in the violet. I 

 moved the prism, and they disappeared ; but when the prism was brought back to 

 its former position, they also returned. I then observed the prism in open light, 

 and saw that it had veins, chiefly opaque and white, running through it, and that 

 there were several of these in the place where the light passed when the prism was 

 held as before. But in case the inclination and shape of these images might be 

 owing to the irregular order in which the veins were laid, I held another prism, 

 which happened to have parallel veins ; in many positions of this the fringes or 

 images returned, not indeed always so regular nor always of the same kind ; for 

 some were confused and broader, formed, as I concluded from this and their posi- 

 tion, by reflexion ; others, made by transparent veins and air-bubbles, were also 

 irregular, but inclined to the red, the violet being farthest from the perpendicular, 

 and these were obviously caused by refraction ; yet all agreed in this, that they were 

 broadest in the red, and narrowest in the violet parts. 



Observ. 5. I held, in the direct rays of the sun, at -f an inch from the small hole 

 in the window-shutter, a glass tube, free from scratches and opaque veins, but like 

 most glass that is not finely wrought, having its surface of a structure somewhat 

 fibrous. When this tube was slowly introduced into the light, and so held that 

 none of the rays might be refracted, a streak, chiefly white, was seen, similar in 

 shape and position to those described in the former paper. When narrowly in- 

 spected, it was found to contain many images by reflexion in it. But these were 

 much diluted by the abundance of white light, reflected without decomposition in 

 the manner above-mentioned. This streak lay wholly on one side of the tube ; but 

 I moved the tube onward a little, and another streak darted through the shadow, 

 and extended all round on both sides : and now, when the tube was in the middle 

 of the rays, there were 1 streaks on both sides, one a little separated from the 

 other and continued through the shadow, the other on each side of the shadow ; 

 the former was evidently produced by refraction ; it contained many images very like 

 those by reflexion, only more vivid in the colours, which were all in the inverted 

 order, the violet being outermost, and the rest nearest the point of incidence. 

 Images similar to these are also producible on the retina, as mentioned before. 



Observ. 6. I now placed a prism at the hole, and made the same images by re- 

 fraction, out of homogeneal light. These inclined to the red, not, like images by 

 reflexion, to the violet ; but they were broadest in the red, and narrower towards 

 the violet parts. In short, when viewed beside the images by reflexion, except in 

 point of brightness and inclination, they differed from them in no respect. The 

 first 3 experiments show, that when homogeneal light is reflected, some rays are 

 constantly disposed into larger images than others are, that is, into images more 

 distended in length, though of the same breadth. The 4th experiment shows, that 

 trie same takes place when light is inflected and deflected ; and the last 2 show, 



