separately; an inner lift being measured from the top curb to the bottom of the cup; a middle lift being taken at its total depth minus the depth of the dip; and a bottom lift is less the dip, and the depth to which it is immersed in the tank in order to keep the seal so as to prevent the gas from blowing.

From a knowledge of the pressure thrown by a holder, and the area of the top of the vessel, we can arrive at a close approximation to its weight; and conversely, knowing the weight and area, we can estimate the pressure which a holder will give. All calculations of this character are based on the weight of a cubic foot of water, which is 62.5 lb., from which we know that a column of water exactly 1 foot square and 12 inches high is equal to 62 lb., or a column of the same area 3 inches high is equal to 15-625 lb., or a column 1 inch high is equal to 5.21 lb.

Consequently, if we wish to ascertain the weight of an uncounterbalanced holder, giving a certain pressure, then on multiplying the area of the top in feet by the weight corresponding to a column of water of the height of pressure given, the weight will be arrived at. In order to arrive at the pressure which will be given by a holder of a certain weight, we divide the total weight of the holder by the area in feet, this will give the weight of a square foot; from which the pressure equal to a column of water of that height is obtained.

The following formulæ shortly express the meaning of the above:

Calling W the weight of the holder in pounds,

A

Р

P =

=

A single-lift gas-holder usually gives a pressure of 3 or 4 inches head of water, the amount thrown depending upon the ratio of the depth to the diameter, while each outer lift of a telescopic holder throws a pressure of about 2 inches head of water.

THE

CHAPTER XIV.

THE PRACTICE OF PHOTOMETRY.

HE object of photometry is to measure the amount of light emitted from a luminous body, and this information is obtained by the aid of a well-known law in optics, viz., that light varies inversely as the square of the distance; i.e., that as a body is removed from any source of light, it receives less and less illumination according to the distances to which it is removed; and that if at any two distances we fix the body and square the distances, the illumination the body receives is inversely as these squares. The above facts could be shown to be as stated by means of the following simple experiments.

"Suppose a person placed in a perfectly dark room, and a light (fig. 37) to be in another room, and a hole a foot square to be cut in the door of the dark room at 1 foot from the light, and a screen to be placed close to the hole in the door. A square 12 inches on each side would be illuminated by the light, and the rest of the screen would be dark.

66

'If the screen be moved a foot from the hole in the door, and therefore 2 feet from the light, the illuminated area will be a square 2 feet on each side, or 4 superficial feet. If the screen be placed 1 foot further from the door, i.e., at 2 feet distance from the hole, and 3 feet distance from the light, its illuminated area will be a square 36 inches on each side, and, therefore, containing 9

superficial feet. Similarly, if it be removed 1 foot further still, i.e., 3 feet from the hole and 4 feet from the light, its illuminated area will be a square of 48 inches on the side, which will contain 16 superficial feet, and so on for every additional foot by which it is removed from the light. Now, the number of feet illuminated will be found exactly the same as the squares of the distances at which the screen is placed from the light.

1 = portion of screen illuminated at 1 foot distance.

=

=

66

"So that the area illuminated is found to vary directly as the square of the distance from the light.

"But the information we require to know is how much light falls on a given unit of surface. We know that all the light which passed through the hole fell upon the screen when it was close to the door, at 1 foot distance from the light, and this is the entire quantity of light received by the screen at each of its positions. But the quantity which was concentrated upon 1 foot, at 1 foot distance, was

spread over 4. feet at 2 feet distance, and therefore the quantity upon each foot of screen must be a fourth of the entire light. Again, at 3 feet distance the light spread over 9 feet of the screen, and therefore the quantity upon each foot must be one-ninth of the whole. In the same way at 4 feet distance the light spreads over 16 feet, and, therefore, the light on each foot must be one-sixteenth of the whole, and so on for any given distance. Therefore, in order to find the amount of illumination of a given area, all that is required is to measure the distance of the illuminated body from the light, and to square that distance, and we then obtain an expression of the quantity of light received by the given area as compared with that which it would receive at a distance fixed upon as unity" (Bowditch).

The celebrated Count Rumford devised a method for the estimation of light by means of what is known as the shadow photometer, and made use of the above-mentioned law.

This method was very generally employed for several years, and in order to make the description quite clear we will explain how to use the shadow photometer for comparing the illuminating powers of a lamp and a standard sperm candle.

Place a vertical rod in front of a white screen (fig. 38). Commence by placing the candle and lamp side by side, about a foot in front of the rod. Two shadows of the rod will be then thrown on the screen, one by the candle, and one by the lamp, the latter being the darker of the two. Now what does this indicate? It must be borne in mind that there are two sources of light illuminating the screen, and that each shadow is only a partial shadow. The part of the screen on which the candle shadow falls receives no light from the candle, but it does receive light from the lamp. In the same manner the part of the screen on which the lamp shadow falls receives no light from the