On Refraction. 205 
along the diameter of its bottom fix little marks at a small di- 
- stance from one another; then, through a small hole in the win- 
dow-shutter of a dark chamber, let in a beam of the sun’s light : 
where the beam falls upon the floor, place your basin so that its 
marked diameter may point towards the window, and that the 
beam of light may fall on the mark that is most distant from the 
window: this done, fill the basin with water, and you will observe. 
that the beam which before fell upon the most distant mark, will 
now, by the refractive power of the water, he turned out of its 
straight course, and fall two, three, or more inches nearer the 
centre of the basin.”’ The fallacy of this experiment can easily 
be explained on the same principle as the first. I shall merely 
remark, that when the water is thrown in, we do not see the 
marks at the bottom of the basin, but reflected images of those 
marks floating on the water; and also the beam of light, when 
falling obliquely on the surface of the water, must cause a re- 
flected image, such as an oar would. Therefore, any conclusions 
drawn from such an experiment must prove erroneous. 
A very simple experiment may be made in the following man- 
ner: Cut a square piece of white paper about the size of a half- 
crown, and let it be dipped in a tumbler of clear water: on 
looking at it, it appears as if split into two papers, giving a sim- 
ple but conclusive illustration of these reflected images. 
I shall now say a few words on refraction through concave and 
convex lenses; nor do I see much occasion to enlarge on this 
part of my subject, having already in my paper on Vision, pub- 
lished in one of your former Journals, shown that the cornea 
and not the retina is the true and only seat of vision, and that 
the mind receives its ideas from minute images painted thereon, 
and not from any crooked refractions forming imaginary images 
in the air. Indeed, a person consulting optical writers, and re- 
ferring to their figures explanatory of telescopes with four lenses, 
must suppose !Nature, instead of being simple and uniform in her 
operations, to be fond of all manner of twistings and turnings. 
At the object-glass the rays get the first twist; two more at the 
medium-glasses ; a fourth at the eye-glass; a fifth at the cornea; 
a sixth at the crystalline lens ; aseventh at the vitreous humour ; 
and, if it were necessary, a dexterous optician may twist it round 
his finger. Newton and De Domenis have done nearly as much 
with their two reflections and two refractions in the rainbow. 
For the experiments with lenses, it is necessary to procure a 
glass globe about three inches in diameter, the bull’s eye of a 
magic-lantern, and a concavo-concave lens. Having pasted a 
piece of black cloth in the shape of the letter T on a pane of glass 
at the window, I requested an assistant, when seated opposite, 
to look steadfastly at it; on now looking into his pupil, I per- 
ceived 
