mination measured directly from the lamp at the distance mentioned, owing to the sensitiveness of the eye changing for different colored lights of different strengths and selective radiation. Two lamps showing a wide difference in candle-power as measured on the photometer bar might not differ greatly in lighting distance as measured with a luminometer or candle-foot photometer at a distant point. When operating indoor photometers, the illumination is relatively strong and the eye is then more sensitive to the yellow rays, that is, waves in the vicinity of the D line. When working with a luminometer or candle-foot photometer in the street, the illumination is relatively weak, that is, two or three thousandths of a foot-candle. In this faint light the eye is more sensitive to the green and violet rays; in other words, the sensitiveness has traveled up the scale from the D to the E line. It is a well-known fact that the alternating enclosed arc of, say, 550 watts emits less light than the direct-current enclosed arc of corresponding wattage. Nevertheless the lighting distance as measured by the luminometer shows very little difference in the illuminating powers of the two lamps. It is possible that the composition of the soft core usually employed in one of the electrodes of the alternating arc has a higher selective radiation in the vicinity of the E line, which may account for the discrepancies cited. Aside from difficulties introduced by a change of sensitiveness of the eye, we are obliged to consider the reduction or change in candle-power due to different enclosing-globe densities, and to depreciation with time owing to the deposit on the globe, to say nothing of the variations due to the traveling of the arc and the time-element reduction factor indispensable to compensate for the variations in light due to the peculiarities of the feeding mechanisms of the different lamps. To represent service conditions, I believe that the reduction factor of about 10 per cent should be added for the enclosed arc and 33 1/3 per cent for the open arc. These figures, however, have not been confirmed by illumination tests. You can therefore appreciate that much careful and difficult work will be necessary before these constants can be established. I can see, however, that the results are most encouraging, and that if we keep at the work we shall arrive at something sufficiently stable to standardize.

I wish to present a few curves illustrating some of the points

mentioned, and also to illustrate the reasonableness of selecting some distant point, for example, 250 feet, as a logical basis of comparison for the measurement of lights intended for street illumination.

Figure 1 shows the characteristic distribution of the light from a 6.6-ampere, direct-current, series enclosed arc, a 7.5

COMPARATIVE CANDLE POWER CURVES OF
SERTES ARCS

Chart showing Maximum and Minimum Vaxation
due to wandering of the Arc

ampere, alternating-current, series enclosed arc, a four-ampere luminous arc, and a 9.5-ampere, direct-current, series open arc. The large and small lobes shown after each light represent the variation that may possibly take place due to the traveling of the arc, feeding or otherwise.

Figure 2 represents the illumination from the lamps above mentioned, except that the variations are averaged. By referring

to either chart it will be observed that the rays of light at an angle of approximately 3.5 degrees below the horizontal will strike the observation plane at 300 feet distant, assuming that the lamp is 20 feet above the line of observation. A ray falling at approximately 5.5 degrees below the horizontal will strike the observation plane at approximately 200 feet from the lamp.

Note particularly that the candle-power of any of the lamps undergoes practically no change within the angle mentioned, that is to say from 3.5 to 5.5 degrees, so that the measurement of illumination anywhere from 200 to 300 feet would be in the most useful portion of the curve for street illumination.

Furthermore, referring to Figure 3, it will be noted that the light falling between two degrees below the horizontal and eleven degrees below the horizontal covers a range of street surface

from 100 to approximately 700 feet. When we consider the large area over which this light is spread, also that it represents practically a fixed candle-power, we can neglect measurements at lower angles that cover relatively small areas. In other words, there would be nothing of importance gained by making measurements at intervals of 100 feet ranging from the lamp to 700 feet distances.

I should like to say a word relative to the Colorado Springs situation that has been mentioned. Aside from the question of contract, a demonstration was originally made before the lamps were accepted. Alternating-current lamps of 6.6 amperes were installed in the streets and members of the city council inspected

Charts Showing Distances from Lamp and Angles that Rays of Light will Fall on Observation Levél.

245

Chart Showing Angles of 5,15 125 Degreee below the Horizontal.

Chart Showing Angles of 11° from the Vertical to 2o below the

Horizontal.

FIG. 3

the lights, made tests by standing between the open arcs and the alternating lamps that were to replace them, and were convinced that the latter were giving the better illumination. There is, however, no question but that the 6.6-ampere alternating-current lamp is on the ragged edge as a substitute for the full arc. When it is so substituted it is usually on the basis of a reduction in price, which I believe applied in this instance. In replacing full arcs it is common practice to substitute a 7.5-ampere alternating lamp or the 6.6-ampere direct-current lamp as a fair substitute, particularly if the contract calls for 450watt lamps; each of the lamps mentioned taking from 450. to 480 watts in service. There are times when the open are would have a greater lighting distance than the enclosed

are taking equivalent energy. There are intervals, however, during which the illumination is not nearly so good, so that by introducing the time-element reduction factor to represent service conditions, even the 6.6-ampere alternating lamp makes a good average showing in contrast with the 9.6-ampere open arc. In view of the testimony presented, I was somewhat surprised at the outcome of the Colorado Springs controversy, but I presume there were conditions with which I was not familiar, which may have had some bearing on the settlement of the case.

MR. SPENCER: The ideas that I have on this subject are embodied in the report, but it may be worth while to call your attention to one or two points in connection with the report. The committee worked very faithfully. Every means of rating a lamp and every means of measuring it after a rating had been determined was carefully considered, and the conclusions reached were practically the following:

First-That the rating should be in terms of illumination; that is, of illumination actually received.

Second-That this illumination should be measured at the most useful point, or at a point where it is most needed in street lighting, namely, approximately midway between the lamps as suspended in the street.

Third-That the measurements should be made in the street and not in the laboratory, and that the means of making the measurements should be as simple as possible, so that a determination as to contract obligations could be made without the necessity of calling in experts, and with means that would be within the reach of the smaller companies. Therefore the committee drew up this specification, providing that the measurement should be at a distance of between 200 and 300 feet from the lamp, providing also for an outdoor measurement, either by means of a luminometer or by means of a portable photometer, and providing further for the use of a 16-cp lamp as a standard of comparison, in order that the illumination received from the arc lamp might be compared with an easily obtainable standard unit.

It seems to me that the specification that has been drawn up embodies these ideas so far as is practicable with the means at present at hand for determining illumination in the street.

MR. F. W. WILLCOX (Harrison, N. J.): I regret that I