694 
PHYSICS: NICHOLS AND WILBUR Proc. N. A. S. 
(a) By its color, which is more often green, blue or white than ruddy. 
(6) By its law of decay, which is that of vanishing phosphorescence in-- 
stead of the law of cooling. 
(c) By its spectrum, which is a characteristic broad banded fluorescence 
spectrum often lacking altogether in the red. The bands of these spectra 
are made up of over-lapping components which form series with the usual 
uniform frequency intervals. 
4. Limits of Activity. — The upper limiting temperature (or tempera- 
ture of extinction) differs with various substances. It is, for example, 
about : 
690° C. for calcium oxide, 
740° C. for calcium sulphide, 
940° C. for zinc oxide and pure zinc sulphide. 
The lower limit also varies through a very wide range; e. g., from 
52° C. for aluminum oxide to 
568° C. for the blue-green band of zinc oxide. 
These limiting temperatures are in some cases very sharply defined. 
5. Individual excitation of bands. — The over-lapping bands of the spec- 
trum are separately excited and each has its particular active range of 
temperatures. As these differ, color changes with temperature result. 
In some cases the temperature ranges do not over-lap, so that two or even 
three fluorescences follow one another as the temperature rises. Thus 
zirconium oxide shows a bluish white fluorescence between 76° C. and 
372°, followed by deep red between 440° and 720°. 
6. Kathodo-luminescence at high temperatures. — Many of the substances 
which exhibit flame excitation are also luminescent in the kathode tube, 
and sometimes we find the same bands brought out and the same upper 
temperature of extinction. 
Thus calcium oxide becomes extinct at about 690°, whether excited 
by the hydrogen flame or by kathode bombardment. 
Sometimes the two types of excitation give strikingly different appear- 
ances, as in the case of the synthetic ruby, which shows the same pale 
green band as other forms of aluminum oxide in the flame but the well 
known brilliant red in the vacuum tube. Our observations indicate 
that under kathodic action also the bands are independently excited and 
that bands of shorter wave-length have a higher temperature range: 
Thus the color shift towards violet with heating recorded by Crookes 
and later systematically studied by Wiedemann and Schmidt (/. c.) is 
confirmed and explained. 
7. Excitation by means of X-rays at these temperatures has thus far 
been tried only with a few substances. It was found, however, that 
calcium oxide, calcium carbonate, sidot blende and some of the phos- 
phorescent sulphides which are subject to flame excitation respond also 
