above the hundredth part of an inch in diameter, and having its surface so perfectly flat and so highly polished that I could see in it the same image of the candle, by placing it extremely near the eye. This chip of wax had no doubt received its flatness and its polish from the surface of a seal, and had started into my eye when breaking the seal of a letter. That this reflecting substance was the cause of the image of the candle, cannot admit of a doubt; but the wonder still remains how the images which it formed occupied so mysterious a place as to be seen without the range of vision, and apparently through the head. In order to explain this, let _m_ _n_, Fig. 2, be a lateral view of the eye. The chip of wax was placed at _m_ at the root of the eyelashes, and being nearly in contact with the outer surface of the cornea, the light of the candle, which it reflected, passed very obliquely through the pupil and fell upon the retina somewhere to the left of _n_, very near where the retina terminates; but a ray thus falling obliquely on the retina is seen, in virtue of the law of visible direction already explained, in a line _n_ C perpendicular to the retina at the point near _n_, where the ray fell. Hence the candle was necessarily seen through the head as it were of the observer, and without the range of ordinary vision. The comparative brightness of the reflected image still surprises me; but even this, if the image really was brighter, may be explained by the fact, that it was formed on a part of the retina upon which light had never before fallen, and which may therefore be supposed to be more sensible, than the parts of the membrane in constant use, to luminous impressions. Independent of its interest as an example of the marvellous in vision, the preceding fact may be considered as a proof that the retina retains its power to its very termination near the ciliary processes, and that the law of visible direction holds true even without the range of ordinary vision. It is therefore possible that a reflecting surface favourably placed on the outside of the eye, or that a reflecting surface in the inside of the eye, may cause a luminous image to fall nearly on the extreme margin of the retina, the consequence of which would be, that it would be seen in the back of the head, half way between a vertical and a horizontal line. LETTER III.