AUGUST 25, 1899.] 
discharge of its phosphorescence had been 
temporarily checked by the cold liquid bath, 
while the phosphorescence of the other card, 
which was at 20° C. throughout the experi- 
ment, was being discharged continuously ; 
the latter, therefore, was slightly more ex- 
hausted. 
3. Two ecards similarly coated with 
Balmain’s paint and which had been placed 
away from light for several days were ex- 
posed to the sun; one (‘ A’) in the ordi- 
nary manner, and the other (‘B’) exposed 
while immersed in liquid air. 
The cards were brought to the dark-room, 
and the one (‘A’), that had not been sub- 
jected to the low temperatare, was immedi- 
ately immersed in liquid air. 
Both cards were then taken out of the 
cold bath and allowed to warm up gradu- 
ally to the temperature of the dark room. 
Almost immediately card ‘B’, which had 
been exposed to sunlight while in liquid air, 
showed phosphorescence, appearing com- 
paratively bright by the time the other 
eard (‘A’) became luminous; the latter, 
however, getting rapidly brighter soon 
phosphoresced more strongly than the card 
“B’, which was exposed to the sun while 
immersed in liquid air. 
This experiment was repeated many 
times with always the same result. 
Several times the test card which had 
been exposed to the sun while in liquid air 
showed faint phosphorescence when still in 
the cold bath. On employing an electric 
arc, however, the Balmain’s paint was 
made to phosphoresce much more strongly 
while at the temperature of liquid air than 
when using sunlight. 
On repeating the experiment with two 
stop-watches it was found that the card 
‘B’, exposed to the sun while in liquid 
air, required not more than 5 to 10 seconds 
to show phosphorescence distinctly, when 
allowed to warm up gradually to 20° C. in 
the dark room. The card ‘A’, simply 
SCIENCE. 247 
exposed to the sun and immersed in liquid 
air in the dark, required 40 seconds as the 
average time to show luminosity. 
In order to obtain an approximate value 
for the temperature corresponding to this 
time of 40 seconds, small sticks of wood, 
weighted with lead, were frozen to the test 
card with alcohol and afterwards with 
mercury. The cards were placed in liquid 
air and then allowed to warm up to the 
room temperature exactly as in the experi- 
ments on phosphorescence. The sticks 
frozen with alcohol fell from the test cards 
in 17 seconds, a value found by ten deter- 
minations, showing little variation from the 
mean. The sticks frozen with mercury fell 
from the card in 90 seconds. From this it 
was obvious that the 40 seconds required 
for the phosphorescence to become visible 
in the experiment mentioned above with 
card ‘A’ corresponded to a temperature 
between the melting point of alcohol and 
that of mercury, or a temperature of from 
75° to 100° below zero ©. Only this rough 
approximation of temperature was made be- 
cause a more accurate determination would 
have had no general significance. This 
is evident when it is remembered that 
the phosphorescing power of different 
samples of the same material, the inten- 
sity of the light employed to excite a phos- 
phorescent substance, and the sensitive- 
ness of the eyes of different observers, all of 
which determine temperature values like 
that under consideration, are factors which 
are subject to considerable variation. 
The results obtained in the foregoing ex- 
periments are in agreement with those pub- 
lished in the Comptes rendus by A. and L. 
Lumiere, already referred to. In the ex- 
periments of these investigators the electric 
are was employed to excite phosphorescence 
in sulphides of calcium and zine at normal 
temperatures; the phosphorescence pro- 
duced was found to become invisible when 
these substances were cooled down from 
