278 TERRESTRIAL REFRACTION AND MIRAGE. 
that I have seen on physics. The observer sees the top of the tree (say 
A) by means of a negatively curved ray, and accordingly it appears to 
him to be in the direction of the tangent to the ray where it enters the 
eye; he sees it in the direction shewn. ‘The other points in the tree he 
sees in other directions, and altogether the tree appears with an in- 
verted image below it. He also sees the tree erect in very nearly its 
true position, because the point 4 is seen by two rays, one ray which 
curves down and another ray, which is traversed in an equal time, which 
passes up through the denser layers and thus escapes this extraordinary 
refraction. or this reason mirage is not extremely important from a 
gunnery point of view. We not only see the inverted image below the 
object, but also the object itself in very nearly its true position. Now 
supposing this tree to be standing out against the blue sky, we not only 
see an inverted image of the tree, but also by curved rays a sky image 
on the ground in front of the tree, that is to say, nearer to the observer 
than the tree; and the tree itself appears to be reflected in this sky 
image, which has very much the appearance of water. This illusion of 
water is very much increased for the reason, that the surface of the 
ground being extremely hot there is great convection of heated air up- 
wards from the surface of the ground which gives a trembling appear- 
ance to the air, and this causes the sky image which is seen through 
this trembling air to have the appearance of being ruffled by light waves 
and ripples. Another circumstance also increases the illusion. The 
curved rays very nearly follow the path of rays reflected at a plane 
mirror placed along the surface of the ground, so that objects appear 
as if they really were reflected at the surface, in what appears to be 
water, but is really an image of the sky. ‘This phenomenon is very 
common in Hgypt. In Heypt you very often have a great quantity of 
sand with villages dotted about, and raised to a certain elevation in 
order to escape the periodical inundations of the Nile; the effect to an 
observer then is of a number of raised villages in the midst of a huge 
lake, the lake beginning at a certain distance in front of the observer, 
say 300 or 400 yards, and then appearing to extend in all directions. 
It is also common in Australia and inthe Plains of India; aud in many 
parts of England in the low lying fen districts, and on the shingle 
ranges at Lydd. 
One very curious thing about mirage is that its depends very much 
upon the position of the eye; afew inches in the height of the eye may 
often make all the difference. J remember distinctly at the artillery 
practice ground at Mooltan in the Punjab ata place called Meeran- 
ki-Serai there was a very fine mirage; the bushes that were on the 
plain all appearing like small islands in a lake, and this could only be 
seen on slightly stooping; by stooping about 6 inches you could see it, 
but if you did not stoop there appeared to be no mirage. Inthe fen dis- 
tricts also there is a beautiful mirage, and the other day I understand 
it could only be seen from the top of a particular wall. It depends upon 
the position of the eye with respect to the stratum of maximum density 
whether the phenomenon is very accentuated ornot. J was very much 
interested too in hearing the other day of a very extraordinary effect 
of refraction from my friend Professor Boys, and I hope he will give 
