Incandescent solids maintained at a constant temperature by the action of the electric cur rent would seem to offer possibilities for the construction of a light standard free from the objections to which all standard flames are nec essarily open. Numerous attempts to produce light standards based upon this principle have been made; the most noticeable being that of Violle. Violle's device consisted of a mass of platinum brdught to its melting point by the action of the current. He proposed to define as the standard the light from a square centi meter of the surface of the metal at the tem perature of solidification. The advantages of a standard capable of such rigorous definition was so obvious that the electrical congress as sembled at Paris in 1881 appointed a committee to determine the question of its practicability. Extended experiments at their hands and on the part of the members of the Imperial-Tech nical Institute in Charlottenburg have unfortu nately led to the conclusion that the platinum standard cannot, with the means ordinarily at command, be made to give constant results.
In the meantime studies of the performance of the ordinary incandescent lamp have shown that we have in it, when properly prepared and. handled, a source of light better adapted as a working standard than any other at present available. The carbon filament heated in vacuo by means of a constant current is subject to changes of illumination only as the result of the slow disintegration of the carbon or of gradual loss of vacuum in the bulb surrounding the fila ment. By the use of storage batteries supple mented by the regulation of a resistance in the circuit it is possible to maintain the filament of an incandescent lamp in a state of incandes cence, the constancy of which leaves little to be desired. Given one such lamp, the intensity of which is known or is arbitrarily taken as the unit, it is possible to determine the voltage at which other lamps have the same intensity, so that it is possible to make copies of the original which, whenever subjected to current at the proper voltage, will return to the intensity at which they were when the comparison was made. Thus while it is not possible to cony struct incandescent lamps which shall at a pre scribed voltage give a definite illumination, it is on the other hand possible by comparison of such lamps with a given standard to find the various voltages at which they will have equal intensities.
The use of the incandescent lamp does not fully solve the problem of the standardizing of light sources; for the determination of the lamp from which the copies are to be made must be by reference to some primary standard such as the Hefner lamp. Having once adjusted one incandescent lamp to agreement with the stand ard, however, one may make copies of these which will agree with one another much more closely than any two Hefner lamps or than a single Hefner lamp upon successive trials will agree with itself. If one were to start anew,
with the Hefner lamp as a standard, and make another set of standardized incandescent lamps, these, though highly uniform, might show no better agreement with the former set than that which one can obtain in subsequent trials with the lamp itself. Such variations of the Hefner lamp amount to about 2 per cent.
The province of the art of photometry does not end with the determination of the intensity of the light sent out in a given direction from any source. One of its chief purposes is to enable us to deal definitely and intelligently with the problems of illumination and these are complicated by the fact that, in general, sources of light do not radiate with equal intensity in all directions. The accepted unit of illumina tion among photometricians on the continent of Europe is the lux, which is the illumination re ceived from a source of unit intensity at a dis tance of one metre. The unit source in this definition is the Hefner lamp already described, so placed that the light is received from it in the horizontal plane. The term bougie-metre is also used in speaking of this unit of illumina tion. The term bougie (or candle) in this defi nition is not any of the standard candles al ready described but a hypothetical candle equal to the Hefner. In countries where British measures are still in vogue a unit of illumina tion frequently employed is the candle-foot which is the illumination afforded by a British standard candle placed at a distance of one foot from the illuminated surface.
The distribution of light from the various sources used in artificial lighting is far from uniform and it is necessary, therefore, in order to give a complete description of the lighting power of any source, to determine the intensity of the source as viewed from all possible di rections. In the case of certain sources of light such as the incandescent electric lamp this is easily accomplished by mounting the lamp at the end of the photometer bar in a holder so constructed as to permit of rotation about both a vertical and a horizontal axis without displacing the centre of the lamp from its po sition. In the case of other sources, such as the arc lamp and most flames, which cannot be tipped from the vertical without modifying the distribution of light or interfering with the performance of the lamp, it is necessary to have recourse to a mirror, placed in the axis of the photometer bar, by means of which light can be reflected along the bar. By varying the angle of the mirror the light can thus be viewed from any desired direction. Corrections must of course be made for loss of light at the surface of the mirror.