NOVEMBER 17, 1899.] 
light. In this case the effects are not addi- 
tive, there being reversal. 
To demonstrate conclusively that the 
time factor was the only one, it was ueces- 
sary to secure an illumination independent 
of the electric spark, and of as short dura- 
tion. This was accomplished in the follow- 
ing manner: A disc 30 cms. in diameter 
was furnished with a radial slit one milli- 
meter wide near its periphery, and mounted 
on the shaft of a high speed electric motor. 
A second slit of equal width was arranged 
in a horizontal position close to the rim 
of the dise, in such a position that the 
two slits would be in coincidence once 
in every revolution. This second slit 
was cut in the wall of a vertical chute 
down which a photographie plate could 
be dropped. By means of a large con- 
vex lens of short focus an image of 
the crater of an arc-lamp was thrown on 
the point of coincidence of the slits. The 
intensity of the illumination transmitted by 
the ‘slits when in coincidence was almost 
sufficient to char paper. The motor was 
then set in motion and a plate dropped 
down the chute. On developing this plate 
three images of the slit appeared, not at all 
overexposed, though the plate was the fast- 
est on the market, and the intensity of the 
light while it lasted comparable to that at 
the focus of a burning glass. By measuring 
the distance between the images and the 
vertical distance through which the plate 
had fallen, it was an easy matter to calculate 
the speed of rotation, which was found to 
be 69 revolutions per second, the air fric- 
tion of the dise preventing higher speed. 
The duration of the exposure will be the 
time occupied by the rim in traveling a dis- 
tance equal to the width of the slit or 1 
mm. This was found to be 1/55000 of a 
second about that of the spark. The crucial 
experiment now remained. A second plate 
was dropped and before development was 
exposed to the light of the candle. The im- 
SCIENCE. TOM 
ages of the slit were most beautifully reversed ex- 
cept at the center where the light was too 
intense. A print from this plate is repro- 
duced in Fig. 5. It seems then that we are 
justified in assuming that the action of an in- 
tense light on a plate for a very brief time inter- 
val decreases the sensitiveness of the plate to light. 
It is curious to contrast with this effect the 
fact that exposure to a dim light for a mo- 
ment or two appears to increase the sensi- 
bility by doing the small amount of pre- 
liminary work on the molecules, which 
seems to be necessary before any change 
can be effected that will respond to the de- 
veloper. I am not prepared to say what 
the nature of the change effected by the 
flash is. Possibly some one familiar with 
the theory of sensitive emulsions can answer 
the question. I have tried using polarized 
light for the reversing flash, and then fog- 
ging one-half of the plate with light polar- 
ized in thesame plane, and other half with 
light polarized at right angles to it. As was 
to be expected there was no difference in the 
CuCOES R. W. Woop. 
PHYSICAL LABORATORY OF THE 
UNIVERSITY OF WISCONSIN, 
MADIson, Oct. 20, 1899. 
ARCHITECTURAL PLANS FOR THE UNIVER- 
SITY OF CALIFORNIA. 
Tuer Phoebe A. Hearst International Com- 
petition for an Architectural plan for the 
University of California was closed on Sep- 
tember 7th by the awarding of five prizes 
for the best plans. The first prize was 
awarded to M. E. Bénard, of Paris; the 
second to Messrs. Howells, Stokes & Horn- 
bostel, of New York; the third to Messrs. 
Despradelle & Codman, of Boston; the 
fourth to Messrs. Howard & Cauldwell, of 
New York, and the fifth to Messrs. Lord, 
Hewlett & Hull, of New York. 
From the outset of their inspection, the 
judges were attracted to the drawings which 
proved, after the awards had been made, to 
