a 
FEBRUARY 15, 1884.] 
light from the Edison per unit of energy as 
compared with the Maxim. 
The results of these comparisons in nine dif- 
_ ferent positions make it possible to establish 
certain comparison equations, from which 
means may be obtained which will serve to 
eliminate, to some extent, the errors of ex- 
periment. 
Let a = the light from the Maxim lamp 
‘edgewise;’ then, by working through the 
different positions of the Edison, the above re- 
sults give — 
a 
2.39a@ 2.717Ta@ 
3.50@ 3.97 
a 
2.717Ta 
Maxim ( v 
t 4 3.580 
and for means — . 
Maxim a 2.64@ 
By a similar computation it is found that— 
_ 4.58a 4.45a 4.360 
Edison LE 3.96a 3.93a 4.38a 
| 3.74a 3.58a 3.864 
the means of which give 
; — 4,.46a 
Edison / 4.08a 
| -3.73a@ 
Fie. 1. Fie. 2. 
i 
| 
| SY 
= | wy 
A 
| 3.73 
x 
i 
Edison. 
. <y 
aS yy 
at 3.48 
INE 2 : 
ee \ 
Maxim. 
Figs. 1 and 2 show the arrangement of these 
intensities of illumination around the carbon 
filament; the plane of the filament being verti- 
eal, and Maxim edgewise being taken as unity. 
For the mean all round, the result is — 
Edison = 4.09 Maxim = 2.44 
4.09 e | 
7 1.676 = 57 in light. 
But from the previous table, 
Le 
1.396 — M iD energy : 
1.676 He ke 
therefore 1.336 > Zor uM” light per elec- 
No 
trical horse-power. 
It seems evident that this difference of 
twenty-five per cent in favor of the Edison 
lamp is largely due to the form of the incan- 
descent filament as compared with that of the 
Maxim lamp. The latter shows great inequal- 
ity in illumination in different directions, the 
light measured from the flat side being about 
three and one-half times as great as that ob- 
SCIENCE. 
185 
tained when the lamp is edgewise. The effect 
of this increased radiating surface is shown 
in the last column of the above table, from 
which it appears, that in the comparison of 
the Maxim, ‘ flat,’ with the Edison in all posi- 
tions, the former shows a higher actual effi- 
ciency than the latter. If this large radiating 
surface could be made to distribute its effect 
around the circumference, the lamp would, in 
the opinion of many, be greatly improved. It 
is fair to say, however, that the unequal dis- 
tribution of light is claimed, by at least some 
of the representatives of this lamp, to be an 
important advantage. It was not so considered 
by the jury. 
The form of the carbon filament in the Edi- 
son lamp is such that a much greater uniformity 
of illumination results. While the Maxim form 
has the advantage of concentrating the radiat- 
ing surface, the arrangement of the carbon to 
accomplish this greatly diminishes its effective- 
ness in the ‘edgewise’ position. Inthe Edison 
there is but a single loop ; and, furthermore, this 
is generally curved, so that it does not lie in 
one plane. As a result, one side of the loop 
never exactly hides the other, and there is but 
little loss from that source. It will be seen in 
the above figure that the illuminating-power 
of the lamp edgewise actually exceeded that 
in any other direction. This difference was 
too constant and too great to be attributed to 
error in experiment. It is attributable, no 
doubt, to the fact, that in this position the lu- 
minous lines lie nearly in the axis of the pear- 
shaped glass containing them, as viewed from 
the photometer-box; there being, therefore, 
less scattering of the light in transmission, and 
possibly some gain on account of reflection. 
Of course, if a lamp were used in which one 
of the branches of the loop exactly or nearly 
covered the other in this position, a different 
ratio of illumination might follow. 
Throughout the entire series of tests the 
jury was fortunate in having the assistance of 
Mr. A. L. Rohrer, a student in physics in the 
Ohio state university. 
In the distribution of work, Mr. Eddy and 
Mr. Laidlaw made the observations, and kept 
the records of the dynamometer work; Mr. 
Laidlaw also taking and reducing the indicator- 
cards. Mr. French made the readings of the 
position of the photometer-box, and set the 
same. Mr. Mendenhall generally read one of 
the galvanometers, and Mr. Rohrer the other ; 
the latter generally keeping the notes of the 
electrical work, although this was done on 
several occasions by Mr. French and by Mr. 
Laidlaw. 
