May 10, 1901.] 
fog it, if they are of sufficient intensity. 
The breadth of the fogged line is 
b= 2a b=2 ¢t. tan a - 
where ¢ is the thickness of the glass plate 
and @ is the critical angle for glass and the 
film substance. Using plates of different 
thicknesses, it was found that, in every 
instance, 6 varied as ¢ to the degree of 
accuracy with which the measurements 
could be made. When a clear glass plate 
was laid on the dry plate, scratching the 
upper plate gave a band whose width varied 
as the combined thickness of both plates. 
If, however, the dry plate were turned over 
so as to have the film between the plates, 
the breadth of the band varied as the 
thickness of the upper plate. And in this 
case when the plate was placed in the 
developer, the band appeared first on the 
outer or upper side of the film. 
It was found that fluorescence does not 
always occur when a diamond is drawn 
over a dry plate. Different diamonds and 
different brands of plates (glass) gave dif- 
ferent results. To determine the cause of 
this apparent irregularity I made a series of 
observations, using ten different diamonds 
and nine kinds of plates. Seven of the 
diamonds were ordinary glass cutters, rang- 
SCIENCE. 
733 
cutter No. 7. Nos. 8, 9, and 10 were Bra- 
zilian black diamonds (in emery wheel cut- 
ters) known as carbonadoes. They ap- 
peared to be amorphous. 
. Ordinary plates, even the most sensitive, 
were not very sensitive to this fluorescence. 
Cramer, Stanley, Seed, Carbutt and Ham- 
mer plates were tried. Sometimes the dia- 
mond produced a slight fogging; usually 
it did not. In this respect no constant 
difference was noted in the different brands 
of plates. 
To determine whether the quality or na- 
ture of glass influences fluorescence I asked 
the M. A. Seed Dry Plate Co. to supply me 
with plates of different kinds of glass coated 
with the same emulsion. The Company 
informed me that they were using four 
brands of glass, English, French, white 
crystal and American, and they sent me a 
supply of each kind coated with their most 
sensitive orthochromatic emulsion. 
Their orthochromatic plate (No. 27) was 
found to be much more sensitive to dia- 
mond-glass fluorescence than the most sen- 
sitive ordinary plate. Fogging almost al- 
ways occurred along the diamond trace, and 
the dark band was frequently very dense. 
The following table embodies the results 
of the tests of different glass with different 
diamonds : 
TABLE I. 
Diamond. Seed’s orthochromatic plates, No. 27. ney niet 
English glass. American. French. White Crystal. “Gilt Edge.” 
1. Small trace 1 | weak 2 | trace 1 | weak 2 | absent 0O 6 
25 we strong 4 | absent 0 | absent 0 | weak 2 | absent 0 6 
See strong 4 | weak 2 trace 1 | fairly strong 3 | trace il) ati 
4, Medium strong 4 | fairly strong 3 | weak 2 | fairly strong 3 | weak 2; 14 
5. sé trace 1 | trace 1 | absent 0 | weak 2 | absent 0O 4 
6 OY trace 1 | absent 0 | absent O | trace 1 | absent 0 2 
7. Large trace 1 | trace 1 | absent 0 | absent 0 | absent 0 2 
8. Carbonado| absent 0 | absent 0 | absent 0 | absent 0 | absent 0O 0 
9. us absent 0 | absent 0:| absent 0 | absent 0 | absent 0 0 
10. a absent 0 | absent 0 | absent 0 | absent 0 | absent 0 0 
Total 16 9 4 13 3 
ing in size from a very small stone in No. 1 
Representing by numbers the relative in- 
(see table) toa large stone in plate glass tensities of the fogged bands, any total in 
