medus, at the seige of Syracuse, burned the ships of Marcellus, by a machine composed of mirrors?

Tutor. Yes: but we have no certain accounts that may be implicitly relied on. Mr. Buffon, about fifty or sixty years ago, burned a plank at the distance of seventy feet, with forty plain mirrors.

James. I do not see how they can act as burning glasses.

Tutor. A plain mirror reflects the light and heat coming from the sun, and will illuminate and heat any substance on which they are thrown, in the same manner as if the sun shone upon it. Two mirrors will reflect on it a double quantity of heat; and if 40 or 100 mirrors could be so placed as to reflect from each the heat coming from the sun, on any particular substance, they would increase the heat 40 or 100 times.

CONVERSATION XI.

Of Concave Mirrors---their Uses--how they act.

JAMES. To what uses are concave mirrors applied?

Tutor. They are chiefly used in reflecting telescopes; that is, in, tellescopes adapted to viewing the heavenly bodies. And as you like to look at Jupiter's little moons and Saturn's ring through my tellescope, it may be worth your while to take some pains to know by what means this pleasure is afforded you.

Charles. I shall not object to any attention necessary to comprehend the principles on which these instruments are formed.

Tutor. A B (Plate 11. Fig. 16.) represents a concave mirror, and a b, c d, e f, three parallel rays of light falling upon it. c is the centre of concavity, that is, one leg of your compasses being placed on c, and then open them to the length c d, and the other leg will touch the mirror A B in all its parts.

James. Then all the lines drawn from c to the glass will be equal to one another, as c b, c d, and cƒ?

Tutor. They will: and there is another property belonging to them; they are all perpendicular to the glass in the parts where they touch.

Charles. That is c b, and c fare perpendi→ cular to the glass at b and ƒ, as well as c d at d.

Tutor. Yes, they are:-c d is an incident ray, but as it passes through the centre of concavity, it will be reflected back in the same line, that is, as it makes no angle of incidence, so there will be no angle of reflection: a b is an incident ray, and I want to know what will be the direction of the reflected ray?

Charles. Since c b is perpendicular to the glass at b, the angle of incidence is abc; and as the angle of reflection is always equal to the angle of incidence, I must make another angle, as c b m, equal to a b c,* and then the line bm is that in which the incident ray will move after reflection.

Tutor. Can you, James, tell me how to find the line in which the incident ray e f will move after reflection?

James. Yes: I will make the angle cfm equal to cfe, and the line fm will be that in which the reflected ray will move; therefore efis reflected to the same point m as a b was.

Tutor. If, instead of two incident rays, any number were drawn parallel to c d, they would every one be reflected to the same point m; and that point which is called the

* To make an angle c b m, equal to another given one, as a b c. From b as a centre with any radius 6 x describe the arc xo, which will cut c b in z, take the distance x z in your compasses, and set off with it z o, and then draw the line bo m, and the angle m c is equal to the angle a b c.

focus of parallel rays is distant from the mirror equal to half the radius c d.

James. Then we may easily find the point without the trouble of drawing the angles merely by dividing the radius of concavity into two equal parts.

Tutor. You may.-The rays, as we have already observed, which proceed from any point of a celestial object may be esteemed parallel at the earth, and therefore the image of that point will be formed at m,

Charles. Do you mean that all the rays flowing from a point of a star, and falling upon such a mirror, will be reflected to the point m, where the image of the star will appear?

Tutor. I do, if there be any thing at the point m to receive the image.

James. Will not the same rule hold with regard to terrestrial objects?

Tutor. No: for the rays which proceed from any terrestrial object, however remote, cannot be esteemed strictly parallel, they therefore come diverging; and will not be