236 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1907. 
The following description, however, shows that Miss Warner and 
Mr. Powrie have solved this problem in a very ingenious way: 
The surface of a plate of ordinary glass is thoroughly cleaned and 
coated with a weak solution of gelatin, albumen, a mixture of the 
two, or of any suitable colloid body containing a proportion of 
alkaline bichromate. The mixture is very similar to the bichro- 
matised fish glue employed by photo-engravers. This coating having 
dried, the plate is exposed under a screen ruled with opaque lines 
which are double the width of the transparent spaces between them. 
The spaces correspond, as we shall see, to the exact width of the 
green and red bands in the manufactured screen. ‘Those portions 
of the sensitized coating which are protected by the lines of the 
screen are not affected by the exposure to light, but the portions 
underneath the spaces in the screen are rendered insoluble on ex- 
posure. “ Development ” takes place in warm water, the mixture 
of glue and albumen dissolves in the unaffected parts, and there 
remains on the plate an enormous number of transparent lines in 
relief separated by depressions which are bare glass. The plates 
are then immersed in a solution of a green dye which penetrates the 
colloid bands, and forms a screen of microscopic green lines. The 
plate is then placed in a bath of alum or tannic acid, which fixes 
the color and enables the bands to attain sufficient intensity. Emerg- 
ing from this bath, the plate is washed and recoated with a sensitive 
mixture and again exposed under the same screen, but with an adjust- 
ment of the carrier in which it rests to such an extent that the green 
lines just produced are protected by the double-width lines of the 
negative. Between the portion of each band left uncovered and those 
stained green there is thus formed a narrow region which is equally 
protected by the opaque bands of the negative. On exposure being 
completed, the plate is passed as before into warm water to develop 
the image, and it is then seen that a screen has been formed with a 
series of green lines and of transparent lines in relief separated by 
intervals narrower than the lines in relief. 
The plate is then plunged into a red dye bath and fixed and mor- 
danted as before. Examined by transmitted light, the screen then 
presents a yellow color, due to the mixture of the red and green rays 
of the two lines which are printed at this stage. It has now to re- 
ceive a third coating of bichromated mixture, and is then exposed 
through the back without the interposition of any negative. The 
light thus reaches the sensitive film through all portions of the 
screen not occupied by green and red lines. As soon as exposure 
is completed it is again developed in warm water and transferred 
to a blue dye bath, which stains only those portions other than the 
red and green, and forms with them a continuous series of colors 
over the whole of the screen. As a result, the screen examined under 
